SAFETY ENHANCED INNOVATIONS FOR OLDER ROAD USERS. EUROPEAN COMMISSION EIGHTH FRAMEWORK PROGRAMME HORIZON 2020 GA No

Transcription

1 SAFETY ENHANCED INNOVATIONS FOR OLDER ROAD USERS EUROPEAN COMMISSION EIGHTH FRAMEWORK PROGRAMME HORIZON 2020 GA No Deliverable No. 4.1b Deliverable Title Dissemination level Draft Test and Assessment Procedures for current and advanced Passive VRU Safety Systems Public Authors Zander, Oliver Hynd, David BASt TRL Approved by Wisch, Marcus BASt 28/05/2018 Issue date 31/05/2018 The research leading to the results of this work has received funding from the European Union's Horizon 2020 research and innovation programme under grant agreement No

2 EXECUTIVE SUMMARY The demographic change European society is facing during the next decades will be challenging for passive vehicle safety. The external road user safety branch of the HORIZON 2020 SENIORS project will address special safety needs in particular of the elderly by defining equivalent safety requirements compared to average age pedestrians within current test and assessment procedures. This will be, alongside new and revised test tools, to obtain appropriate assessment of the vehicle protection potential, also taking into account the ongoing changes in injury patterns of vulnerable road users since the introduction of consumer test programmes and regulative requirements. Deliverable D4.1(b) of the SENIORS project summarises the history of pedestrian safety, starting with the early biomechanical programmes and describing the development of test procedures for different body regions using impactors until the currently applied state of the art test and assessment procedures within legilation and consumer test programmes. A study of recent collision scenarios alongside injury patterns of pedestrians and cyclists reveals the coverage of actual needs by the current procedures and identifies open injury protection gaps. New and modified test and assessment procedures for the most relevant body regions will address those needs, leading into a draft rating scheme for Box 3 (Vulnerable Road Users) of the European New Car Assessment Programme (Euro NCAP) which will, in the end, synthesise the particular assessment tools to an overall rating for vulnerable road users (VRU). The developed test and assessment procedures will serve for physical tests whose results are reported in Deliverable D4.2(b). The following partners contributed to this deliverable report: Partner Representative Chapters BASt Oliver Zander 1-6 TRL David Hynd Page 2 out of 79

3 CONTENTS Executive summary Introduction The EU Project SENIORS Background and Objectives of this Deliverable Literature review History of pedestrian testing Full scale tests and first steps towards impactor test procedures EEVC Working Group EEVC Working Group European Directive 2003/102/EC State of the art of pedestrian test and assessment procedures Legislation Regulation (EC) No. 78/ United Nations Regulation Summary of possibilities to obtain European pedestrian type approval Consumer Test Programmes Early years of Euro NCAP Today s Euro NCAP Test and Assessment Procedures Lower or Upper Legform to Bumper Test Upper Legform to WAD 775 Test Headform to Bonnet / Windscreen / A-Pillar Test Headform to Bonnet Leading Edge Test Active Pedestrian Protection Systems Heavy Commercial Vehicles Euro NCAP Box 3 Rating Other NCAP s Overview Collision studies and identification of open injury protection gaps Up to date injury patterns including the elderly Coverage by current and future test and assessment procedures Updated test and assessment procedures Headform to bonnet leading edge/bonnet/windscreen Test Tool Test Procedure Derivation of Impactor Thresholds Test Synthesis TIPT to bonnet leading edge/bonnet/windscreen Test Tool Test Procedure Derivation of Impactor Thresholds Test Synthesis FlexPLI-UBM to bumper/bonnet leading edge Test Tool Test Procedure Derivation of Impactor Thresholds Femur Tibia Page 3 out of 79

4 Knee Test Synthesis Draft Euro NCAP Box 3 rating scheme Summary and Conclusions Conclusions and impact on SENIORS project Transposition of results...69 Glossary References Acknowledgements Disclaimer Page 4 out of 79

5 1 INTRODUCTION 1.1 THE EU PROJECT SENIORS Because society is aging demographically and obesity is becoming more prevalent, the SENIORS (Safety ENhanced Innovations for Older Road users) project aims to improve the safe mobility of the elderly, and overweight/obese persons, using an integrated approach that covers the main modes of transport as well as the specific requirements of this vulnerable road user group. This project primarily investigates and assesses the injury reduction in road traffic crashes that can be achieved through innovative and suitable tools, test and assessment procedures, as well as safety systems in the area of passive vehicle safety. The goal is to reduce the numbers of fatally and seriously injured older road users for both major groups: car occupants and external road users (pedestrians, cyclists, e-bike riders). Implemented in a project structure, the SENIORS project consists of four technical Work Packages (WP1 WP4) which interact and will provide the substantial knowledge needed throughout the project. These WPs are: WP1: Accidentology and behaviour of elderly in road traffic WP2: Biomechanics WP3: Test tool development WP4: Current protection and impact of new safety systems In addition, there is one Work Package assigned for the Dissemination and Exploitation (WP5) as well as one Work Package for the Project Management (WP6). 1.2 BACKGROUND AND OBJECTIVES OF THIS DELIVERABLE The demographic change European society is facing during the next decades will be challenging also for passive vehicle safety. For more than two decades, test and assessment procedures for passive vehicle safety systems have been developed and applied for the benefit of road participants as either car occupants or external road users being involved in a collision with passenger cars. On the one hand, these test and assessment methods set a Page 5 out of 79

6 standard towards vehicle type approval regarding their passive safety. On the other hand, consumer programmes such as the European New Car Assessment Programme (Euro NCAP) established a milestone in terms of vehicle safety exceeding the regulatory requirements while giving a detailed overview of the vehicle safety performance to the costumer. The external road user safety branch of SENIORS addresses special safety needs in particular of the elderly by defining equivalent safety requirements to passenger cars within test and assessment procedures, alongside the provision of new and revised test tools, towards an appropriate assessment of the vehicle protection potential. It thereby takes into account the ongoing changes in injury patterns of vulnerable road users. In depth accident analyses as reported in SENIORS Deliverable 1.2 (Wisch et al., 2017) revealed in several cases a change in injury patterns alongside the demographic change our society is facing. From these changes, Hynd et al. (2017) derived a variety of new biofidelity requirements for the older external road users such as pedestrians and cyclists, as reported in SENIORS Deliverable 2.1. A revision of existing test tools and pedestrian impactors aimed at implementing those new requirements into updated or, where necessary, new tools, as reported in Deliverable 2.5b (Zander et al., 2018). After the tool validation reported in Deliverable 3.2b, (Zander et al., ) the test and assessment procedures were updated, incorporating the new tools and impactors. Based on a review and evaluation of the current test and assessment procedures, this Deliverable proposes updates, where necessary, incorporating the new test tools and impactors with the updated biomechanical limits. Deliverable 4.1(b) of the SENIORS project summarises the history of pedestrian safety and investigates the recent injury patterns of pedestrians and cyclists, also having a focus on the elderly. The state of the art test and assessment procedures within legislation and consumer test programmes are compared to current injury patterns of pedestrians and cyclists involved in accidents with passenger cars. Page 6 out of 79

7 Identified actual needs will be covered by a set of new and modified test and assessment procedures for the most relevant body regions, leading into a draft rating scheme for Box 3 of Euro NCAP. Impactor threshold values will be derived from studies of impact biomechanics reported in Deliverable 2.5b alongside correlation studies between impactor simulations and human body model simulations as described in SENIORS Deliverable 3.2b. A synthesis of the particular assessments will be proposed as an example overall rating for vulnerable road users (VRU) within Euro NCAP. 2 LITERATURE REVIEW In order to identify possible open gaps in terms of further needs of vulnerable road users towards vehicle based protection during vehicle to VRU accidents, a study of pedestrian safety history until currently applied test and assessment procedures and subsequent comparison to current accident scenarios and injury patterns is necessary as a first step. 2.1 HISTORY OF PEDESTRIAN TESTING Full scale tests and first steps towards impactor test procedures For many years, unprotected road users like pedestrians have accounted for a significant proportion of casualties on European roads. In the late 1970s, pedestrian collision statistics investigated the vehicle front as a major problem for pedestrian injuries. In this context, the first biomechanics programmes for the protection of pedestrians (both, children as well as adults), funded by the European Commission, were defined, focusing on vehicle velocities up to 40 km/h. Full scale tests were carried out with post mortem human subjects (PMHS) as well as crash test dummies (see Figure 1) against different vehicle front shapes. While the first proposal for a test procedure was defined in 1982 (compare Figure 2), the repeatability of the head impact kinematics was identified as one of the major problems of a full-scale procedure. Page 7 out of 79

8 Figure 1: Full-scale test according to 1981 test procedure. Figure 2: Full-scale test procedure with pedestrian dummy, as proposed in 1982 (Faerber et al., 1982). Thus, further research focused on component testing. In 1987, the EEC ad-hoc working group 'Erga Safety' discussed a proposal to assess pedestrian safety by using impactor tests, prepared by the Department of Transport of the United Kingdom and based on a development from the European Enhanced Vehicle-safety Committee EEVC, formerly known as European Experimental Vehicles Committee (Commission of the European Communities, 1986). The proposal contained component tests with pedestrian impactors (horizontal legform tests to the vehicle s bumper, horizontal tests to the bonnet leading edge using a hipform and tests perpendicular to the bonnet top with child and adult headform impactors) see Figure 3. Page 8 out of 79

9 Figure 3: First developments of legform (left) and hipform (right) impactors for pedestrian component testing (European Economic Community, 1985) EEVC Working Group 10 This proposal for an EC Directive was found very promising; however, EEVC saw the need for further research to close identified injury protection gaps. Therefore, in 1987 the EEVC Working Group 10 Pedestrian Protection was established. The work of the group concluded in a final report dated 1994, proposing a set of subsystem test methods with the impactors illustrated in Figure 4, comprising of a legform to bumper test, an upper legform to bonnet leading edge test and a headform to bonnet top test (EEVC, 1994). Possible challenges for car manufacturers in getting to the required safety levels were approached by proposing a phased-in introduction of the requirements. Page 9 out of 79

10 Figure 4: Pedestrian legform, upper legform and headform impactors according to EEVC WG 10 (EEVC, 1994) EEVC Working Group 17 From 1997 on, EEVC Working Group 17 Pedestrian Safety as successor of WG 10 was tasked with a review of the proposed test methods and incorporating possible modifications under consideration of new and existing collision data. A final report with revised test methods and impactors was issued in 1998 and updated in 2002 (EEVC, 2002). The main changes in comparison to the previously existing methods were: the incorporation of a damping element into the lower legform impactor to avoid vibrations during the impact; the introduction of a horizontal test with the upper legform against vehicles with high bumpers; a capping of impact energies for the upper legform to bonnet leading edge test; a change of headform impactor material for avoiding vibrations and increased durability; an improvement of certification methods; and modified impact areas for the headform tests. Page 10 out of 79

11 An overview of pedestrian impactors as defined by EEVC WG 17 is illustrated in Figure 5; the corresponding pedestrian component tests are summarised in Figure 6. Figure 5: Pedestrian legform, upper legform and headform impactors according to EEVC WG 17 (EEVC, 2002). Upper legform impactor Child headform impactor Adult headform impactor Upper legform impactor Legform impactor Oliver Zander, BASt Figure 6: Test methods according to EEVC WG 17 (Zander, 2017). Page 11 out of 79

12 2.1.4 European Directive 2003/102/EC The tests according to EEVC WG 17 formed the basis for the first European legislation in the field of pedestrian safety. European Directive 2003/102/EC was planned to be introduced in two phases. The first phase had to be mandatorily applied to new passenger cars for the purpose of European Whole Vehicle Type Approval as from October 2005 onwards. While Phase 1 was introduced with sometimes less stringent impactor thresholds and testing the windscreen and the bonnet leading edge for monitoring purposes only, Phase 2 was planned to be an entire mapping of the EEVC WG 17 test procedures and requirements. Figure 7 summarises the test methods according to the European Directive. Head: 3,5 kg, 35 km/h HPC 1000 (2/3 of area) HPC 2000 (1/3 of area) mm on bonnet, 50 (WSS - monitoring, 35 km/h, 4,8 kg, 35 ) Lower Leg: 40 km/h, 200g, 21, 6 mm Upper Leg: SUV, Bumper - 40 km/h, 7,5 kn, 510 Nm (on BLE - monitoring, km/h, 5 kn, 300 Nm) Head: 2,5/4,8 kg, 40 km/h HPC 1000 (entire area) Child: mm on bonnet, 50 Adult: mm on bonnet, 65 Lower Leg: 40 km/h, 150 g, 15, 6 mm Upper Leg: on BLE km/h, 5 kn, 300 Nm SUV, Bumper - 40 km/h, 5 kn, 300 Nm or appropriate measures! Figure 7: Test methods according to European Directive 2003/102/EC (Zander, 2017). In parallel to Phase 1, a separate Directive related to the use of frontal protection systems ( bullbars ) prescribed construction requirements on the one hand and performance requirements for frontal protection systems as original equipment or to be sold as separate technical units on the other hand (European Union, 2005). Page 12 out of 79

13 2.2 STATE OF THE ART OF PEDESTRIAN TEST AND ASSESSMENT PROCEDURES While Phase 1 of the European Directive on pedestrian protection was entirely introduced as intended, Article 5 prescribed a study to be performed on the feasibility of the Phase 2 requirements for European Car Manufacturers. Several studies, mainly a feasibility study on behalf of the Commission (Lawrence et al., 2004), a study on the technical feasibility issued by the Japan Automobile Research Institute (JARI, 2004) and an equal effectiveness study on pedestrian protection (Hannawald et al., 2004) in parallel to discussions between the EC and car makers finally led to a revised Phase 2 with less stringent requirements, provided that a Brake Assist System was to be introduced for all new vehicle types. The revision, realised as Regulation (EC) No. 78/1009, furthermore contained the introduction of modified headform impactors with a mass of 3,5 kg (child) and 4,5 kg (adult), both with a diameter of 165 mm, the cease of monitoring phases for the (EEVC WG 17) adult headform to windscreen test and, as from 24 February 2014 also for the upper legform to bonnet leading edge test, furthermore the reduction of head impact speeds and the introduction of relaxation zones on the bumper test area. Besides, the prescriptions for frontal protection systems were incorporated as well Legislation Regulation (EC) No. 78/2009 The recent Regulation (EC) No. 78/2009 describes the up to this day valid pedestrian protection tests that need to be performed for European Vehicle Type Approval, referring to the technical requirements according to Commission Regulation (EC) No 631/2009. The pedestrian protection tests consist of the following subsystem tests: - Child (50 ) and adult (65 ) headform to bonnet at 35 km/h impact speed: a minimum of 2*9 tests - EEVC lower legform impactor to bumper at 40 km/h impact speed: a minimum of 3 tests Page 13 out of 79

14 - Upper legform impactor to bumper at 40 km/h impact speed, replacing the lower legform test, for high bumper vehicles only: a minimum of 3 tests - Upper legform to bonnet leading edge at varying impact speeds and angles, for monitoring purposes only (requirement to provide the results ceased in 2014): a minimum of 3 tests The tests with the lower or upper legform to the bumper are to be carried out one each to the middle and to the outer thirds of the bumper at positions most likely to cause injuries, at a minimum distance of 132 mm between each other and at a minimum distance of 66 mm inboard of the bumper corners. The tests with the upper legform to the bonnet leading edge are to be carried out one each to the middle and to the outer thirds of the bumper at positions most likely to cause injuries, at a minimum distance of 150 mm between each other and at a minimum distance of 75 mm inboard of the corner reference points. The tests with the child and adult headform impactors are both to be carried out three each to the middle and three each to the outer thirds of the bonnet test area limited by two lines 82.5 mm inboard of the side reference lines, the wrap around distance (WAD) 1000 or 82.5 mm rearward to the bonnet leading edge reference line, whichever line is most rearward, and WAD 2100 or 82.5 mm forward to the bonnet rear reference line, whichever line is most forward. Tests with the child headform impactor have to be always located 165 mm apart from each other and forward to WAD Tests with the adult headform impactor have to be always located 165 mm apart from each other and rearward to WAD The bonnet area itself has to be divided in two performance zones, one of them comprising two thirds of the test area (HPC 1000 zone) and the other one the remaining third (HPC 1700 zone), as illustrated in Figure 8: Page 14 out of 79

15 Figure 8: Determination of head impact zones for European vehicle type approval (European Union, ). In addition, frontal protection systems (FPS) marketed as separate technical units need to be submitted to the following subsystem tests: - EEVC lower legform impactor to FPS at 40 km/h impact speed; - Upper legform impactor to FPS at 40 km/h impact speed, replacing the lower legform test, for high FPS only; - Upper legform impactor to FPS leading edge at varying impact speeds and angles, for monitoring purposes only; - Child / small adult headform impactor to FPS, for wrap around distances (WAD) greater than 900 mm only. Finally, Regulation (EC) No 78/2009 in conjunction with Commission regulation (EC) No 631/2009 lays down detailed provisions for Brake Assist Systems (BAS) which need to be included in all new vehicle types marketed in the European Union since November A summary of the test methods and requirements according to Regulation (EC) No. 78/2009 in conjunction with Commission Regulation (EC) No. 631/2009 is depicted in Figure 9. Page 15 out of 79

16 Figure 9: Test methods and requirements according to Regulation (EC) No. 78/2009 (Zander, 2017). In parallel to the introduction of Regulation (EC) No. 78/2009 with its technical prescriptions, Commission Regulation (EC) No. 631/2009, the European Directive 2003/102/EC and the Directive 2005/77/EC as well as their technical prescriptions 2004/90/EC and 2006/368/EC, were repealed United Nations Regulation 127 Besides the European Regulation, a Regulation on Pedestrian Safety under the Agreement from 1958 of the United Nations was established and entered into force in late Since that point in time, UN Regulation No. 127 (UNECE, 2013) can be applied for pedestrian vehicle type approval in the framework of the European Whole Vehicle Type Approval EU-WVTA (European Union, 2007). United Nations Regulation UN-R 127 is the transition of United Nations Global Technical Regulation UN-GTR9 under the 1998 agreement into national law. Alternatively to Regulation (EC) No 78/2009, UN-R 127 can be applied for European Type Approval. The latest series of amendments, UN-R needs to be Page 16 out of 79

17 mandatory applied since 31 December The main differences between UN-R 127 and Regulation (EC) No 78/2009 are as follows: - Upper legform to bonnet leading edge tests are not foreseen; - Type approval of frontal protection systems is not included; - Type approval of BAS is not included, but needs to be performed according to the requirements described in UN-R 13H (UNECE, 2014); - As from the first series of amendments onwards, the lower legform to bumper tests are to be performed with the Flexible Pedestrian Legform Impactor (FlexPLI) (UNECE, 2015); and - As from the second series of amendments, the bumper test area is modified, using the wider of the two areas limited by the ends of the bumper beam and the corners of bumper, latter ones defined by contact points with corner gauges (UNECE, 2016). Page 17 out of 79

18 Summary of possibilities to obtain European pedestrian type approval Figure 10 summarises the different possibilities for European pedestrian type approval: 2007/46/EC 78/ /2009 UN-R 127 & UN-R 13H 2005/ /368 Phase 1 Phase 2 as from all M1 & N1 2,5 t as from all M1 and N1 UN-R (GTR9 Phase 1 ) as from UN-R FlexPLI, New head test area determination asfrom UN-R BTA as from Figure 10: Possibilities for pedestrian type approval according to Directive 2007/46/EC (Zander, ) Consumer Test Programmes Early years of Euro NCAP Prior to the introduction of pedestrian test procedures within European legislation, the European New Car Assessment Programme (Euro NCAP) developed an assessment and rating scheme for the pedestrian protection potential of passenger cars that was applied for the first time in The test procedures, starting with three child and adult headform tests, three upper legform and three lower legform tests was subjected to many revisions, modifications and updates. With respect to the assessment, the standalone pedestrian rating scheme with a maximum of four stars was incorporated into an overall rating in Since that point in time, a good overall rating of vehicles was depending on the degree of fulfilment in each of the Page 18 out of 79

19 four boxes adult occupant protection, child safety, pedestrian protection and safety assist according to the philosophy of weighting and balancing at the same time Today s Euro NCAP Test and Assessment Procedures In the subsequent year, a Euro NCAP Working Group on Pedestrian Safety started working on the improvement of pedestrian test and assessment procedures towards enhanced passive pedestrian safety, which were stepwise implemented into the test and assessment protocols between 2012 and 2015 (Zander et al., 2015). The tests that are performed in the Euro NCAP programme are as follows: Lower or Upper Legform to Bumper Test Since 2014, the pedestrian protection potential of the bumper area of passenger cars is assessed by performing impactor tests with the flexible pedestrian impactor FlexPLI. Different to regulatory procedures, the test markup is done using a grid method with grid points at a distance of 100 mm apart from each other and located between the corners of bumper or the ends of the bumper beam, whichever area is wider. Different to UN-R , the corners of bumper are defined by planes contacting the vehicle outer contour under an angle of 60 degrees to the vertical longitudinal vehicle centreplane. Every second grid point at either the left or right hand side of the vehicle is to be tested. Car manufacturers are given the opportunity to additionally nominate grid points that otherwise would not be tested. The grid points that are not tested are allocated the worse of the two adjacent results. The FlexPLI is propelled horizontally, at an impact speed of 40 km/h and an impact height of 75 mm above ground level against the vehicle frontend, see Figure 11. Page 19 out of 79

20 Test area: bumper Impactor: FlexPLI Impact speed: 40 km/h Impact angle: 0 Impact height: 75 mm above GL Figure 11: FlexPLI to bumper test procedure The score for each grid point is calculated by balancing the injury risks related to the medial collateral ligament (MCL) and the tibia segments, whereas points for the MCL are only awarded in case of not exceeding the identified risk of 10 mm elongation for cruciate ligament rupture (ACL/PCL). As far as the tibia is concerned, only the highest of the four bending moments is taken into account for the assessment. Between the defined upper and lower performance limits for MCL elongation and tibia bending moment, a sliding scale is applied to both criteria. The total score for the lower legform area is calculated by adding the points for the individual grid points and scaling the results to 6 points. For visualization, a five colour scheme is applied. Figure 12 summarises the FlexPLI test and assessment procedure: Calculationofgrid pointscore - Tibia BM & MCL (use 10 mm ACL/PCL switch) - Use of sliding scales - Weighting of tibia and knee (50%/50%) Score ( GP) = Score ( Tibia BM max) + Score ( MCL) Calculation of total score scaled to 6 points n i= 1 Score ( GPi ) *6 n Visualization Figure 12: FlexPLI test and assessment procedure according to Euro NCAP (Zander et al., 2015). Page 20 out of 79

21 In case of the lower bumper reference line at the point(s) to be tested is between 425 and 500 mm above ground level, at the choice of the vehicle manufacturer the legform tests may be alternatively performed by using the upper legform impactor (so called high bumper test ), as illustrated in Figure 13. If the lower bumper reference line at the point(s) to be tested is more than 500 mm above ground level, the tests shall be performed with the upper legform impactor. Test area: bumper Impactor: Upper legform Impact speed: 40 km/h Impact angle: 0 Figure 13: High bumper test procedure Upper Legform to WAD 775 Test Some years ago, Euro NCAP fundamentally changed the upper legform test method, moving away from focusing on the bonnet leading edge as a potentially injury causing vehicle part towards the upper leg as the injured body region. Therefore, since the year 2015, the upper legform test is performed with the impactor centreline aiming at wrap around distance 775 mm. Similar to the FlexPLI and thus different to regulatory procedures, the markup is done using a grid method with grid points at a distance of 100 mm apart from each other and located between the corner reference points. Again, every second grid point at either the left or right hand side of the vehicle is to be tested. Car manufacturers are given the opportunity to additionally nominate grid points that otherwise would not be tested. The grid points that are not tested are allocated the worse of the two adjacent results. Page 21 out of 79

22 Impact velocity and impact energy are calculated using the impact angle, latter one measured between the direction of impact and the horizontal. The direction of impact is determined from the velocity vector which is perpendicular to a connection line between two points on the internal bumper reference line and WAD 930, both located on the identical vertical longitudinal plane as the impact point: Test area: WAD 775 Impactor: Upper legform v c = v 0 cos(1.2α) / v 0 = 11,1 m/s E n = 0,5 * m n * v c ² / m n = 7,4 kg v t = [2 * E n / 10,5 kg] 0,5 Figure 14: Upper legform to WAD775 test. For scoring, no weighting of the maximum bending moments and the maximum sum of forces is done, but, as also before the fundamental changes to the test procedure, the worst of the two results is taken into account only. For the allocation of points to the particular test results, a sliding scale is applied between the upper and lower performance limits, which in terms of the maximum bending moments was lowered to 285 or 350 Nm respectively. The total score for the upper legform area is calculated by adding the points for the individual grid points and scaling the results to 6 points. For visualization, the five colour scheme is applied similar to that of the FlexPLI. Figure 15 illustrates the principles of the upper legform to WAD 775 test and assessment procedure. Page 22 out of 79

23 Calculation of grid point score - Femur BM & Σ Femur Forces - Use of sliding scales - Worst of the results taken Score ( GP) = Min[ Score ( Femur BM max); Score (( Femur F) max )] Calculation of total score scaled to 6 points n i= 1 Score ( GPi ) *6 n Visualization Figure 15: Upper legform to WAD 775 test and assessment procedure according to Euro NCAP (Zander et al., 2015) Headform to Bonnet / Windscreen / A-Pillar Test Tests with the ISO adult and child headform impactors are performed to the vehicle frontend at an impact speed of 40 km/h within an area limited by the wrap around distances WAD 1000 and WAD 2100 and the side reference lines. Since 2013 already the selection of test points is done randomly on points of a headform grid with a resolution of 100mm * 100mm, located between the side reference lines, WAD 1000 and WAD 2100 with a transition zone between WAD 1500 WAD Tests in this zone that are located on the bonnet are to be performed with the child headform impactor; grid points on the windscreen are to be tested with the adult headform impactor, see Figure 16. Page 23 out of 79

24 Test area: WAD Transitional zone: WAD Impactor: Child & Adult headform Impact speed: 40 km/h Impact angle: 50 (CH) / 65 (AH) Oliver Zander, BASt Figure 16: Headform to bonnet / windscreen / A-Pillar test After the provision of the safety performance by means of colour information for each of the determined grid points by the OEM, a default minimum of 10 and, upon request of the vehicle manufacturer, a maximum of 20 verification points is generated randomly. Additionally, the manufacturer is given the opportunity to select a maximum of 8 blue zones, each consisting of 1 or 2 grid points, where the safety performance is unknown or previous tests have indicated instable results, and which are tested once on the point selected by Euro NCAP, unless symmetries are being applied. Thus, in total a minimum of 10 and a maximum of 28 head impact tests are to be performed according to the headform grid method. After testing, a correction factor is calculated by the quotient of the sum of actual verification test results and the sum of the points resulting from the colour predictions. The correction factor provides an indication of how precisely actual testing matches the prediction and should be between 0.75 and 1.25 for the predictions to be accepted by Euro NCAP. By multiplying the sum of points obtained by the colour predictions times the correction factor, the actual performance of the predicted grid points is calculated. In a last step, the total head score is calculated by including the number of green defaulted points and the actual test result from the blue zones, and scaled to the 24 points that are available for the head performance in box 3 (pedestrian protection) of the Euro NCAP overall rating scheme. A visualization of the test results by applying a five colour code scheme defined by the underlying results to each individual grid point completes the head assessment. The principles of the headform to bonnet / windscreen / A-Pillar test and assessment procedure according to Euro NCAP is summarised in Figure 17. Page 24 out of 79

25 10-20 Verification tests 8Tests in blue zones Calculation of correction factor Actual Score CF = Predicted Score (Verification Points) (Verification Points) Calculation of total score scaled to 24 points s ( Score( SGPi ))* CF + DGGP + Score( BGPi ) i 1 1, = = i Total Score scaled = *24 SGP + DGGP + DRGP + BGP b SGP: Standard Grid Point CF: Correction Factor DGGP: Default Green Grid Point DRGP: Default Red Grid Point BGP: Blue Grid Point Visualization Figure 17: Headform to bonnet / windscreen / A-Pillar test and assessment procedure according to Euro NCAP (Zander et al., 2015) Headform to Bonnet Leading Edge Test Alongside the change of the upper legform test in 2015, Euro NCAP introduced a monitoring test with the ISO child headform impactor to the bonnet leading edge, see Figure 18. The reason was that with the introduction of the new upper legform test, the bonnet leading edge was not anymore in the focus of Euro NCAP passive pedestrian safety assessment. On the other hand, Longhitano et al. (2005) and Zander (2014) found the bonnet leading edge having a high relevance as an injury causing part in real world collision data. While in most cases, the bonnet leading edge is to some extent also covered by the upper legform to WAD 775 test, in case of being located between WAD 930 and WAD 1000 the bonnet leading edge reference line (BLE-RL) is not assessed anymore. Therefore, this test is only performed in case of the bonnet leading edge reference line is located between WAD 930 (representing the rearward limitation of the upper legform test area) and WAD 1000 (representing the forward limitation of the headform test area) (Euro NCAP, 2014). This test is performed at an impact angle of 20 to the horizontal. Page 25 out of 79

26 Test area: BLE reference line Impactor: Child headform Impact speed: 40 km/h Impact angle: 20 Figure 18: Headform to bonnet leading edge test Active Pedestrian Protection Systems Since several years, active systems of passive pedestrian safety have been introduced into the market. In particular, active bonnets which provide additional clearance between bonnet inner panel and the underlying structure are expected to give additional benefit to the protection of pedestrians in case of an impact. Since legal requirements do not address several important aspects in terms of an assessment of the actual safety performance of active bonnets, Euro NCAP has developed a test protocol, including prerequisites for bonnets to be tested in the active (deployed) state. These prerequisites contain the protection below the deployment threshold, the correct bonnet deployment timing, the detection of all pedestrian statures, the bonnet deflection due to upper body contact and the protection at higher impact speeds (Zander, ) Heavy Commercial Vehicles M1 vehicles derived from commercial vehicles with a gross vehicle mass between 2,5t and 3,5t and between 8 and 9 seats (including the driver) have been tested for a couple of years according to a different Euro NCAP procedure, mainly containing deviating impact angles for the headform impactors and omitting the upper legform to BLE test for vehicles with a BLE height greater than 835 mm. Since the pedestrian test conditions for these heavy commercial vehicles have been incorporated into the pedestrian testing protocol, the only remaining difference between these vehicles and Page 26 out of 79

27 passenger cars is a homogeneous head impact angle of 50 rearwards to the bonnet leading edge reference line of commercial vehicles. Thus, the requirements for heavy vehicles were deleted from the heavy vehicles protocol and a reference to the pedestrian testing protocol inserted (Euro NCAP, 2015) Euro NCAP Box 3 Rating As described in the previous chapters, based on the work of the pedestrian working group, several fundamental changes have been undertaken by Euro NCAP in terms of passive pedestrian testing and assessment of the pedestrian protection potential of motor vehicles, predominantly passenger cars. However, the weighting factors between the body regions head, upper and lower leg have remained unchanged at 24/36 (66,6%), 6/36 (16,6%) and 6/36 (16,6%). Since the introduction of AEB pedestrian systems in 2016 and AEB cyclist systems in 2018, the portion of passive pedestrian safety within Box 3 has been lowered to 36/42 (85,7%) in 2016 and 36/48 (75%) in AEB systems on the other hand are only taken into account in case of a minimum passive pedestrian safety performance of 22 points (Euro NCAP, ). For the Euro NCAP overall rating, the minimum fulfilment rates in Box 3 as summarised in Table 1 are needed as balancing criteria. Table 1: Box 3 weighting, in-box balancing and overall balancing of pedestrian safety performance in Euro NCAP. The weighting of Box 3 w.r.t. the overall rating remains at 20%. Page 27 out of 79

28 Other NCAP s Besides Euro NCAP, further consumer test programmes have been introduced during the last two decades, some of them already with implemented pedestrian test and assessment procedures, such as the Japan New Car Assessment Programme (JNCAP) or the Korean New Car Assessment Programme (KNCAP), which are in many areas very alike to Euro NCAP. While following the approach of vehicle categorisation for the head impact area for a long time, JNCAP has decided in the meanwhile to follow the Euro NCAP approach, introducing the grid method for the head as well as the lower leg area. For latter ones, the FlexPLI was introduced with weighting factors differing to those of Euro NCAP, with a portion of 73% for the tibia and 27% for MCL, using ACL/PCL as an MCL switch at 13 mm. FlexPLI thresholds and sliding scales differ to those of Euro NCAP ( Nm for tibia bending moment, mm for MCL elongation). The new upper legform test to WAD 775 has not been introduced yet. KNCAP mainly follows the procedures of Euro NCAP with headform and FlexPLI tests. Different to Euro NCAP, no transitional area for child and adult headform tests is foreseen. The assessment of tests with the FlexPLI is done according to Euro NCAP. As within Euro NCAP, for high bumper vehicles, the upper legform is used instead of the FlexPLI. In the US it was planned to introduce a consumer pedestrian test and assessment protocol very similar to the procedures of Euro NCAP, including FlexPLI, the grid method as well as upper legform to WAD 775 and the headform to BLE test (NHTSA, 2015). However, since the 2017 election and in the course of installing a new administration in NHTSA, the process has been put on hold. Page 28 out of 79

29 2.2.3 Overview Table 2 summarises the current pedestrian component test parameters and requirements according to European legislation as well as the Euro NCAP programme: Table 2: European pedestrian safety standards and biomechanical limits (Zander, 2017). Page 29 out of 79

30 3 COLLISION STUDIES AND IDENTIFICATION OF OPEN INJURY PROTECTION GAPS 3.1 UP TO DATE INJURY PATTERNS INCLUDING THE ELDERLY In depth collision analyses revealed in several cases a change in injury patterns alongside with the demographic change our society is facing. Figure 19 shows the general change in injury patterns of pedestrians and cyclists. While head and lower leg signed responsible for both groups of vulnerable road users for the majority of AIS 2+ and AIS 3+ injuries in collisions on German roads with passenger cars registered between 1995 and 2005, thoracic and, in terms of pedestrians, also pelvic injuries gained more importance in the last years. Figure 19: Injury patterns occurring in vehicle to pedestrian/cyclist collisions according to AIS98 code based on body parts. The Figure shows the relations between the injury severity from AIS 2 (or AIS 3 respectively) to AIS6 (i.e. AIS 2+ or AIS 3+ respectively) in the particular body region and AIS 2+ (AIS 3+) injury severities of all body regions (Zander et al., ). A study of German and Swedish crash data from GIDAS and STRADA, as reported in SENIORS Deliverable 1.2 (Wisch et al., and Wisch et al., ), concluded the mostly affected body regions for pedestrians and cyclists and their particular relevance for the elderly. Showing that for pedestrians head, thorax, pelvis and lower extremities remain the most relevant body regions with the highest portions of AIS 2 and AIS 3+ injuries, with the elderly suffering more frequently from severe injuries than younger pedestrians, bicyclists show the highest injury levels for Page 30 out of 79

31 the head, the thorax and the lower extremities being the key affected body regions for both age groups. GIDAS (n=916) STRADA (n=3443) Figure 20: Percentages of injury severities for the different pedestrian body regions within GIDAS and STRADA. Each column adds up to 100 percent by adding all percentages from AIS0 to AIS9. (Wisch et al., 2017). In-depth investigations of the German GIDAS and Swedish STRADA data show the thorax representing a higher percentage of severe injuries for both groups of vulnerable road users, pedestrians and cyclists. Meanwhile, the importance of head and lower extremity injuries is remaining in most cases at the same level as before. Furthermore, injury severities of the elderly as vulnerable road users especially in the described body regions are higher than for the age group between years. It thus can be concluded, that the main focus in the revision and further development of impactors and test procedures needs to be settled to the head, the thorax and the lower extremities. Page 31 out of 79

32 When focusing on the elderly (65+), Wisch et al. (2017) found the most frequently affected body regions of pedestrians and cyclists in crash databases and hospital databases as follows: Table 3: Most frequently affected body regions (mais2+) of older road users (65 years and more) by road user type identified in crash databases (Wisch et al., 2017). Ranking (mais2+) Cyclist Pedestrian 1 Thorax Lower Extremities 2 Upper Extremities Head 3 Lower extremities Upper Extremities Table 4: Most frequently affected body regions (mais2+) of older road users (65 years and more) by road user type identified in hospital databases (Wisch et al., 2017). Ranking (mais2+) Cyclist Pedestrian 1 Head Head 2 Thorax Pelvis+lower extrem 3 Pelvis+lower extrem Thorax Table 5: Most frequently affected body regions (mais3+) of older road users (65 years and more) by road user type identified in crash databases (Wisch et al., 2017). Ranking (mais3+) Cyclist Pedestrian 1 Head Lower Extremities 2 Thorax Head 3 Lower extremities Thorax Table 6: Most frequently affected body regions (mais3+) of older road users (65 years and more) by road user type identified in hospital databases (Wisch et al., 2017). Ranking (mais3+) Cyclist Pedestrian 1 Head Head 2 Thorax Thorax 3 Lower extremities Pelvis+lower extrem Head, thorax and lower extremities are thus confirmed to be the mostly affected body regions, with the upper extremities also relevant for mais2+ within in depth accident databases. This, however, was not confirmed by the available hospital data where head, thorax lower extremities and pelvis were the always predominant body regions. Page 32 out of 79

33 3.2 COVERAGE BY CURRENT AND FUTURE TEST AND ASSESSMENT PROCEDURES A detailed study of pedestrian and cyclist injuries on injury level for both, pedestrians and cyclists, further details the relevance of the different body regions: Table 7: Pedestrian injuries in GIDAS with indication of the relevance of particular body regions for different injury levels all accidents with M1 vehicles as from MY 2006 onwards; known gender and ages (Study performed by Vukovic, 2017). Pedestrian Injuries in GIDAS (for all age groups) All AIS1+ AIS2+ AIS3+ Body Region (AIS08REG) n % n % n % n % 0 - Other 5 0,4% 0 0,0% 0 0,0% 0 0,0% 1 - HEAD ,6% ,5% 67 19,5% 32 31,1% 2 - FACE 106 7,7% 106 7,8% 8 2,3% 0 0,0% 3 - NECK 7 0,5% 7 0,5% 1 0,3% 1 1,0% 4 - THORAX 93 6,7% 90 6,6% 42 12,2% 31 30,1% 5 - ABDOMEN 52 3,8% 48 3,5% 8 2,3% 1 1,0% 6 - Cervical Spine 64 4,6% 63 4,6% 36 10,5% 6 5,8% 7 - Upper Extremities ,6% ,9% 35 10,2% 0 0,0% 8 - Lower Extremities ,0% ,4% ,7% 32 31,1% 9 - Unknown 2 0,1% 2 0,1% 0 0,0% 0 0,0% Total ,0% ,0% ,0% ,0% Pedestrian Injuries in GIDAS (for 65+ age group) All AIS1+ AIS2+ AIS3+ Body Region (AIS08REG) n % n % n % n % 0 - Other 0 0,0% 0 0,0% 0 0,0% 0 0,0% 1 - HEAD 63 13,9% 60 13,5% 15 10,1% 8 16,7% 2 - FACE 29 6,4% 29 6,5% 6 4,0% 0 0,0% 3 - NECK 2 0,4% 2 0,4% 1 0,7% 1 2,1% 4 - THORAX 38 8,4% 36 8,1% 20 13,4% 14 29,2% 5 - ABDOMEN 11 2,4% 10 2,2% 3 2,0% 0 0,0% 6 - Cervical Spine 30 6,6% 29 6,5% 23 15,4% 6 12,5% 7 - Upper Extremities 99 21,8% 99 22,2% 15 10,1% 0 0,0% 8 - Lower Extremities ,9% ,4% 66 44,3% 19 39,6% 9 - Unknown 1 0,2% 1 0,2% 0 0,0% 0 0,0% Total ,0% ,0% ,0% ,0% Page 33 out of 79

34 Table 8: Cyclist injuries in GIDAS with indication of the relevance for of particular body regions for different injury levels - all accidents with M1 vehicles as from MY 2006 onwards; known gender and ages (Study performed by Vukovic, 2017). Cyclist Injuries in GIDAS (for all age groups) All AIS1+ AIS2+ AIS3+ Body Region (AIS08REG) n % n % n % n % 0 - Other 11 0,3% 0 0,0% 0 0,0% 0 0,0% 1 - HEAD 314 8,9% 306 8,8% 53 15,6% 23 26,7% 2 - FACE 329 9,3% 327 9,4% 15 4,4% 1 1,2% 3 - NECK 16 0,5% 16 0,5% 4 1,2% 3 3,5% 4 - THORAX 183 5,2% 178 5,1% 47 13,8% 28 32,6% 5 - ABDOMEN 63 1,8% 60 1,7% 10 2,9% 0 0,0% 6 - Cervical Spine 144 4,1% 139 4,0% 19 5,6% 5 5,8% 7 - Upper Extremities ,5% ,9% ,8% 3 3,5% 8 - Lower Extremities ,9% ,2% 84 24,7% 23 26,7% 9 - Unknown 20 0,6% 14 0,4% 0 0,0% 0 0,0% Total ,0% ,0% ,0% ,0% Cyclist Injuries in GIDAS (for 65+ age group) All AIS1+ AIS2+ AIS3+ Body Region (AIS08REG) n % n % n % n % 0 - Other 0 0,0% 0 0,0% 0 0,0% 0 0,0% 1 - HEAD 61 11,6% 57 11,1% 15 18,1% 6 20,7% 2 - FACE 61 11,6% 59 11,5% 5 6,0% 1 3,4% 3 - NECK 1 0,2% 1 0,2% 1 1,2% 1 3,4% 4 - THORAX 46 8,8% 45 8,8% 17 20,5% 9 31,0% 5 - ABDOMEN 3 0,6% 2 0,4% 1 1,2% 0 0,0% 6 - Cervical Spine 13 2,5% 12 2,3% 3 3,6% 2 6,9% 7 - Upper Extremities ,2% ,9% 23 27,7% 3 10,3% 8 - Lower Extremities ,2% ,5% 18 21,7% 7 24,1% 9 - Unknown 2 0,4% 2 0,4% 0 0,0% 0 0,0% Total ,0% ,0% ,0% ,0% The GIDAS collision data provides first indications regarding a possible new weighting within the Euro NCAP pedestrian rating. When including both vulnerable road user groups, setting the focus on AIS2+ injuries would result in weighting factors between the body regions head, thorax and lower extremities at 27%, 20% and 53% including all age groups and 20%, 25% and 55% for the elderly (65+). Herewith it can be concluded an almost equal significance of body regions regardless the age. Focusing on AIS3+ injuries, almost identical relevance for all three body regions can be stated over all age groups, with thoracic injuries predominating very slightly (35%). For the elderly, AIS3+ injuries to the lower extremities are predominating (41%), followed by thorax (36%) and head (22%). All evaluations of the collision data underline that the current Euro NCAP test and assessment procedures don t reflect Page 34 out of 79

35 the actual collision situation on German roads, overrepresenting the head (67%) and underrepresenting the lower extremities (33%). 4 UPDATED TEST AND ASSESSMENT PROCEDURES As reported in SENIORS Deliverable 2.5b (Zander et al., 2018), due to first indications from collision investigations the efforts for bringing forward passive pedestrian safety were focused on the body regions head, thorax and lower extremities. During simulations against the generic SAE Buck and its derivatives it was found that an improvement of the impactor by adding an additional neck mass did not contribute to a better impact kinematics, eliminating a sufficient portion of impactor rotation and thus unrealistic readings of linear accelerations. It was decided at an early point in the SENIORS project to concentrate further efforts in terms of revisions and further impactor developments on the test tool Thorax Injury Prediction Tool (TIPT) for assessing thoracic injuries, as well as a significant adaptation of the flexible pedestrian legform impactor (FlexPLI) by adding an upper body surrogate representing the pedestrian s and cyclist s torso mass. Subsequent to the validation of these two impactors as reported in SENIORS Deliverable 3.2b (Zander et al., ), the test and assessment procedures for performing physical tests need to be updated, incorporating the new tools, test parameters and ambient conditions. 4.1 HEADFORM TO BONNET LEADING EDGE/BONNET/WINDSCREEN Test Tool A study on developing a headform impactor with additional neck mass was undertaken already within the European FP6 research project APROSYS (Advanced Protection Systems). Brüll et al. (2009) reported about first good results with improved impactor kinematics and time histories of the resultant headform accelerations getting closer to those obtained during human body model simulations. The intention within SENIORS was to follow up this development and undertake further modifications and improvement to the new head neck impactor (HNI), based on simulations with both, human body models as well as the HNI on generic vehicle frontends. However, it was not possible to demonstrate the superior behaviour and Page 35 out of 79

36 benefits of the HNI on the two dimensional frontend. Finally, it was decided by the SENIORS consortium to retain the design of the headform impactor and to focus on impact parameters for including cyclists within the existing test procedures toward a better degree of protection of all vulnerable road users Test Procedure Based on findings from collision research, human body model simulations and tests with the child and adult headform impactors, Zander et al. (2017) developed a combined vulnerable road user test procedure. For that purpose, the existing pedestrian headform test procedure was merged with a standalone bicyclist test procedure, while not compromising the protection of pedestrians, offering the best possible protection also for bicyclists during an accident. While the new procedure would mean some significant changes to legislation, e.g. an increased impact speed of 40 km/h and the inclusion of the windscreen area up to WAD 2500 or the windscreen rear reference line into the assessment (see Figure 21), the main changes to the current Euro NCAP test procedures are a rearward extension of the head impact zone until always WAD 2500 and a moderate adaptation of head impact angles on the windscreen (70 degrees to the horizontal) to better address the safety needs and impact conditions of cyclists. An overview of the impact areas, test conditions and requirements is given in Table 9 and Figure 22. Figure 21: Combined VRU test procedure for legislation (Zander et al., 2017). Page 36 out of 79

37 Table 9: Requirements (upper performance level UPL and lower performance level LPL) for the combined Euro NCAP vulnerable road user test procedure (Zander et al., 2017). Surface Test area HIC Bonnet (WAD /BRRL) Windscreen (BRRL WAD 2500) Child head area (WAD /BRRL) Child & Adult head area (WAD /BRRL) Child head area ( WAD 1500) Child & Adult head area ( WAD 2500) 650 (UPL) 1700 (LPL) 650 (UPL) 1700 (LPL) 650 (UPL) 1700 (LPL) 650 (UPL) 1700 (LPL) v = 40 km/h WAD WAD > 1700 Child Head (3,5 kg) Adult Head (4,5 kg) WS 70 BRRL 70 WS Bonnet 50 / 20 * * FWD to BLE-RL 65 Bonnet Figure 22: Test conditions for the combined Euro NCAP vulnerable road user test procedure (Zander et al., 2017). Page 37 out of 79

38 4.1.3 Derivation of Impactor Thresholds Since the focus of SENIORS regarding head protection of pedestrians and cyclists was set on an improvement of kinematic correlation between impactor and human towards a limitation of rotation to a more realistic intent and thus the acquisition of realistic linear accelerations for calculation of the head performance criterion (HPC), no new human injury criteria were included into this procedure. Thus, HPC performance limits remain unchanged at 650 and 1700 respectively. A value of 1000 was established for the head injury criterion (HIC) for child and adult dummies for many years, and was proven to be effective in limiting serious injuries to the head (EEVC, 2002). Regarding child head injury criteria, NHTSA evaluated several techniques, including scaling of adult data and accident reconstructions (NHTSA, 1996). It was concluded that no single method or set of data stands out clearly as the best choice, because actual biomechanical data are insufficient and of limited applicability. Therefore, NHTSA recommended a HIC value of 1000 also for the child head. Besides, EEVC WG 17 took the decision to use a head performance criterion (HPC) with 15 ms time window for the HIC calculation in both the child and adult headform tests: HPC = HIC 15 A head performance criterion of 1000 according to NHTSA injury risk curve was to be interpreted as a 50% risk for AIS2+ injury as skull fracture: Page 38 out of 79

39 Figure 23: Injury risk curve for Head Injury Criterion (NHTSA, 1999). The new Euro NCAP upper performance limit for the headform impact (HPC 650) that was introduced with the grid is no longer according to the EEVC WG 17 proposal (HPC 1000), but has been adopted from the Assessment Protocol for Adult Occupant Protection (upper performance limit for HIC 36 ), coming from Prasad and Mertz (SAE Paper and SAE paper ). The Euro NCAP lower performance limit (HPC 1700) was introduced in the course of pedestrian protocol update and harmonisation with the legal requirements where the HPC must not exceed a value of 1700 (compare chapter ) Test Synthesis An update of the current Euro NCAP pedestrian head procedure to a combined VRU procedure mainly refers to a modification of test parameters and a rearward extension of the test area. The grid approach as developed by the Pedestrian Safety Working Group, remains unchanged. For consistency with the current assessment procedure, the total amount of points should be calculated as before by scaling of the total grid point score down to the maximum number of achievable points for this subsystem test. Page 39 out of 79

40 4.2 TIPT TO BONNET LEADING EDGE/BONNET/WINDSCREEN As reported in SENIORS Deliverable 2.5b, the ES2 dummy already performed well during full-scale tests for assessing the actual pedestrian performance of a vehicle with high bonnet leading edge. At a quite early stage of the project it was decided to uncouple the thorax from the ES2 dummy to carry out impactor tests. FE simulations with the Total Human Model for Safety, version 4 (THUMSv4) and the ES2 ribcage on identical impact positions of a generic vehicle frontend showed to a certain extent comparability but also some lack of correlation in terms of rib intrusions. Several simulation loops were performed and further improvements in terms of the thorax injury prediction tool as well as ambient conditions for the tests were found, as reported in Deliverable 3.2b Test Tool As a test tool, the ribcage of the ES2 side impact dummy has been uncoupled and used as standalone thorax injury prediction tool. The latest simulations concluded to add the ES2 arm on the impact side, and perform the tests with the arm in the stowed position, which is realised using a TIPT suit with fixed sleeve. A pusher device has been designed for attachment to the pedestrian test stand, with the possibility of adjustment of the impact angle within the launcher to the values defined according to the test procedures, as described in the following chapter. TIPT Launcher Arm (stowed) Figure 24: Thorax Injury Prediction Tool (TIPT), jacket not shown. Page 40 out of 79

41 4.2.2 Test Procedure For an assessment of the VRU protection potential of passenger cars related to thoracic injuries, information on the human and dummy anthropometry as well as the kinematics of impact is needed. In principle, the entire vehicle front can contain potentially injury causing parts affecting the human thorax. The test and assessment area for the thorax however further depends on the vehicle height and the dimensions of the impactor to be used. Figure 25 summarises the most relevant human data for the 5 th female, the 50 th male, the 95 th male (DIN, 2005) and the thorax-related proportions derived from the human body models THUMSv4 and the family of MAthemtical DYnamic MOdels (MADYMO): THUMSv mm 1308 mm 1485 mm upmost rib lowmost rib Lower end lower rib Sternum lower Sternum upper Oliver Zander Figure 25: Anthropometric data: human, THUMSv4 (50 th ) and MADYMO. Thus, based on anthropometric data (including all statures), the test area should be described by wrap around distances coinciding with the height of the lowermost rib of the six years old (6YO) and the height of the uppermost rib of the 95 th adult male. Page 41 out of 79

42 Under consideration of THUMSv4 50th, the area would then be described by WAD 1192 and WAD Taking into account the MADYMO family (6YO 95 th ), the test area would be limited by WAD 770 and WAD The use of the forward limitation covering the 6YO is expected to cover a broader range of statures than needed for the elderly. It is nevertheless considered since the test procedure is meant to address all vulnerable road users within a holistic approach. Since the TIPT is derived from the ribcage of the ES2 side impact dummy, the test area needs to be verified against all locations on the vehicle front that can be potentially impacted by the ES2. Figure 26 shows the dimensions including the theoretical standing height of the ES2 dummy. 498 mm Lower leg (Floor to Knee joint) 388 mm Upper leg (Knee joint to Hip Joint ) 833 mm Torso Neck and Head (Hip joint) to Top of Head) 1719 mm Theoretical Standing Height Top edge of upper rib to bottom edge of lower = 160mm Rib height = 46 mm Rib pitch= 57 mm 1236 mm mid rib centreplane 1293 mm upper rib centreplane 1179 mm lower rib centreplane 1156 mm 1316 mm standing height of ribcage area Figure 26: Dimensions and theoretical standing height of ribcage area of ES2 dummy. From these measurements, the potential impact area of the ES2 ribcage can be calculated between WAD 1156 and The entire ribcage is thus covered by the impact area described by the anthropometric data as shown in Figure 25. Page 42 out of 79

43 As illustrated in Figure 27 in by: the TIPT test area is thus proposed to be determined WAD 770 (height of lowermost rib of 6YO) WAD 1540 (height of uppermost rib of 95 th ) Side Reference Lines (700 mm straight edge inclined by 45 inwards on both sides of the car) WAD 1540 SRL SRL WAD 770 Figure 27: Test area for TIPT to bonnet leading edge / bonnet / windscreen test. All impact points should be aimed at with the intersection of the TIPT mid rib centreplane and the TIPT vertical rib centreplane. Furthermore, all impact points should be located within the test area and 80 mm rearward to of WAD 770; 80 mm forward to WAD 1540; and mm laterally inwards the SRL. A grid (resolution: mm * 160 mm) should be marked on the vehicle front, starting with the intersection of y0 with the point 80mm rearwards of WAD 770, Page 43 out of 79

44 marking a grid point every 160 mm in rearwards wrap around direction until 80 mm forward to WAD Starting from each y0 intersection with the particular WAD, it should be moved laterally rightwards (leftwards) and a grid point should be marked every mm until mm laterally inwards the SRL, see Figure mm 267mm 133.5mm 80mmm Figure 28: Relation between ES2 ribcage dimensions and TIPT markup. A prediction of grid point results should be given by the vehicle manufacturer for the performance in terms of maximum rib deflection, to be indicated with the colours green yellow orange brown red, as outlined in chapter Random test point selection should be done according to the Euro NCAP headform test procedure (Euro NCAP, 2018). Impact speeds and angles have been investigated in SENIORS Deliverable 2.5b with FE simulations of TUC THUMS against the SAE Buck and its derivatives (Zander et al., 2018). It has been found that thorax speed and impact angle mainly depend on the geometry of the vehicle front. Thus, the TIPT angles and speeds derived from the HBM simulations are defined by the vehicle to be tested. A categorisation of vehicles is done according to a revised procedures, following in principle the method developed by the International Harmonized Research Activity Pedestrian Safety Working Group IHRA (Mizuno Y., 2005). The flowchart in Figure 29 summarises the methodology for vehicle categorisation. Page 44 out of 79

45 Bonnet angle < 20 Y BLE < 835 Y N N Sedan Van/MPV SUV Figure 29: Vehicle categorisation (modified IHRA Method) for determination of impact speed, impact direction and TIPT angle. As illustrated in Figure 30, the frontends of the SAE Buck and its derivatives fit well within the categorisation according to the flowchart: Figure 30: Plausibility check Actual parameters of SAE Buck Subsequent to the categorization of the vehicle frontends, the TIPT impact speeds, TIPT impact angles and angles of the velocity vectors that were derived from TUC THUMS simulations and the corresponding parameters of the thorax are defined as follows: Page 45 out of 79

46 25 / 20 km/h 26 / 30 km/h SEDAN 39 / 40 km/h 28 / 50 km/h Instant of impact (ms) T1-T12 average relative translational velocity *with respect to the car **values in global coordinate system Vx (mm/ ms) Vz (mm/ ms) Angle (º) / 20 km/h 30 / 30 km/h 31 / 40 km/h 32 / 50 km/h Instant of impact (ms) T1-T12 average relative translational velocity *with respect to the car **values in global coordinate system Vx (mm/ ms) Vz (mm/ ms) Angle (º) / 20 km/h 34 / 30 km/h SUV 35 / 40 km/h 36 / 50 km/h Instant of impact (ms) T1-T12 average relative translational velocity *with respect to the car **values in global coordinate system Vx (mm/ ms) Vz (mm/ ms) Angle (º) TIPT impact angle TIPT impact angle Figure 31: Derivation of TIPT impact speeds, angles of velocity and TIPT impact angles. For pragmatic reasons, the TIPT impact angles and the angles of the velocity vector are rounded to integer values. These angles, due to the kinematics of the human thorax during the impact, differ from each other, i.e. that also the angle of the velocity vector in the component test is not perpendicular to the TIPT. An overview of the Page 46 out of 79

47 different angles depending on the corresponding vehicle categories is given in Figure 32. Vehicle category Sedan SUV Van / MPV TIPT impact angle 15 (75 ) 20 (70 ) 28 (62 ) Angle of velocity TIPT impact speed 27 km/h 15 km/h 21 km/h 15,2 12,7 23,8 Sedan SUV Van/MPV 40 km/h 40 km/h 40 km/h Figure 32: TIPT impact parameters Derivation of Impactor Thresholds Until 2014, the ES2 dummy was used as a car occupant during the side and pole side impact within the Euro NCAP programme. Limits for the lateral chest compression were used as assessment indicator for 45 and 67 years old car occupants suffering AIS 3 injuries. While the upper performance limit for maximum rib intrusion in both crash modes was set to a value of 28 mm, representing a 5% AIS 3 injury risk for the 67 YO, the lower performance limit and, as far as the side impact with the mobile deformable barrier is concerned, the capping limit, was defined at 50 mm, indicating a 30% AIS 3 injury risk for the 45 YO. The capping limit for the pole side impact was defined at 55 mm, representing a 50% risk of the 45 YO for suffering AIS 3 injuries (Euro NCAP, 2014). Lowne et al. (n.d.) developed risk curves for suffering AIS 2+, AIS 3+ and AIS4+ thoracic injuries as functions of peak rib deflections out of PMHS tests and Page 47 out of 79

48 normalised the results to a 45 YO and to the production prototype EUROSID injury parameters, see Figure 33. Figure 33: Probability of a thoracic injury of severity AIS3+ with confidence intervals (left) and AIS2+, AIS3+ and AIS4+ without confidence intervals (right). For the assessment of occupant protection during side impacts, Euro NCAP may have assumed that the production prototype and production EUROSID were equivalent (at least at the thorax) and that the design intent of the ES2 was to fix some issues while maintaining biofidelity. Therefore, the applicability of EuroSID injury risk functions was concluded. Since the relevance of thoracic injuries increases with increasing injury severity for both, pedestrians as well as cyclists, independent of the age group (compare Table 7 and Table 8), in light of available data it seems most convenient for the assessment of the protection potential of passenger cars to focus on AIS3+ injuries. To be in line with the 5 colour scheme being applied in Euro NCAP, the allocation of threshold values is proposed according to Table 10. Page 48 out of 79

49 Table 10: TIPT thresholds (maximum rib intrusion). Colour (points) Maximum rib intrusion Interpretation (human injury risk) Green (1) 28mm 5% AIS3 (67 YO) Yellow(0.75) 35mm 20% AIS3+ (45 YO) Orange (0.5) 40 mm 30% AIS3+ (45 YO) Brown (0.25) 44mm 40% AIS3+ (45 YO) Red (0) 49mm 50% AIS3+ (45 YO) Test Synthesis Following the approach for the assessment of the measurements of the upper legform impactor, the minimum score for each impact location is taken into account, only. As for the headform test, a grid approach is also followed for the TIPT impact procedure, see chapter For consistency with the current assessment procedure, the total amount of points will be calculated as before by scaling of the total grid point score down to the maximum number of achievable points for this subsystem test. 4.3 FLEXPLI-UBM TO BUMPER/BONNET LEADING EDGE In SENIORS Deliverable 2.5b, extensive FE simulations with the FlexPLI Baseline, the FlexPLI-UBM rigid and the FlexPLI-UBM rubber were performed against a generic test rig representing a broad variety of vehicle frontends, and compared with human body model simulations with THUMSv4. The results of this programme were validated by simulations against actual vehicle frontends and the SAE Buck and its derivatives SUV and Van/MPV, as reported in SENIORS Deliverable 3.2b. As one of the results, both versions of FlexPLI with applied upper body surrogate showed a kinematic behaviour that was superior to the FlexPLI Baseline, with further advantages of the FlexPLI with UBM attached via flexible element. Furthermore, time histories of the UBM versions were much more in line with the HBM than those of the baseline impactor. Altogether it was decided to further follow a test and assessment Page 49 out of 79

50 procedure using the FlexPLI rubber with a total mass of 20.2 kg which aims at replacing the FlexPLI baseline test as well as the test with the upper legform impactor to WAD 775, as currently performed by Euro NCAP Test Tool The FlexPLI, as implemented within UN-R 127 (UNECE, 2015), is intended to remain except for improvements of robustness of materials unchanged. Additionally, an upper body mass of 6.2kg, representing the torso of a pedestrian, is attached with a flexible rubber element to the upper femur part of the FlexPLI, see Figure 34. Figure 34: FlexPLI with upper body mass with flexible connection (FlexPLI-UBM rubber ). Basic components of the UBM with a total mass of 7.0 kg (including the rubber element and the base) are illustrated in Figure 35. Page 50 out of 79

51 Urethane cover hip/flesh profile and mass protection 2x steel wires on hip rotation point Rubber element Leg support guide Flex PLI top plate UBM Base plate Figure 35: Upper body mass with flexible connection. The intention of a flexible attachment of the upper body mass is to add hip rotation to the impactor kinematics and to simulate the time lag that can be observed in the kinematics of the human body model: Figure 36: Comparison of kinematics between human body model THUMSv4 and FlexPLI- UBM rubber during simulations against a generic test rig (Zander et al., ). Page 51 out of 79

52 The FlexPLI-UBM rubber has a nominal total length of mm, subdivided in a tibia length of 497 mm, a femur length of 431 mm, and a length of the UBM including the flexible attachment of mm. Figure 37 gives an overview of the assembly. 497 mm 431 mm mm Figure 37: Dimensions of measurement locations of FlexPLI-UBM rubber Test Procedure For an assessment of the VRU protection potential of passenger cars related to lower extremity injuries, information on the human and dummy anthropometry as well as the kinematics of impact is needed. In terms of the test area to be assessed with the FlexPLI-UBM rubber, reference is made to the bumper test area as defined by Euro NCAP, i.e. either limited by the corners of bumper identified by contacting the vehicle with vertical planes under an angle of 60 degrees to the vertical longitudinal vehicle centreplane, or by the outer ends of the bumper beam (compare chapter ). Page 52 out of 79

53 Figure 38 summarises the most relevant human anthropometric data for the 50 th male (DIN, 2005) and the lower extremity related proportions derived from THUMSv4. THUMSv4 521 mm 970 mm Figure 38: Anthropometrical data: human and THUMSv4 (50 th ). Thus, based on anthropometric data and the geometries of the FlexPLI, the impact conditions for the FlexPLI-UBM rubber should be defined by the bonnet leading edge (BLE) height as well as the height of the lower bumper reference line (LBRL). According to the anthropometric data from DIN (THUMSv4), an assessment of femur injuries should be undertaken only for vehicles with a minimum height of the BLE exceeding 535 mm (521 mm). Second, the height of the LBRL should not exceed 1045 mm (970 mm), which is always the case with currently available M1 vehicles. According to the geometry of the FlexPLI-UBM rubber, the WAD of the BLE should not be lower than 522 mm since this value marks the lower end of the femur when the FlexPLI rubber being tested at an impact height of 25 mm. Figure 39 summarises the test conditions for the FlexPLI-UBM rubber : Page 53 out of 79

54 WAD 953 mm 522 mm 497 mm 431 mm GL 25 mm Figure 39: Test conditions and impact heights for FlexPLI-UBM rubber. From the previous considerations and possible interactions of the UBM with the vehicle frontend, leading to unintended peaks or unrealistic timings of the waveforms (compare SENIORS Deliverable 3.2b), the prerequisite of the upper femur end to be not lower than the BLE reference line of the vehicle to be tested can be derived. For the impact height of the FlexPLI-UBM rubber, the following parameters are defined: BLE-RL WAD 953: GL + 25 mm 954 BLE-RL WAD 1000: GL + 26 mm 72 mm BLE-RL WAD > 1000: GL + 72 mm Examples of vehicles with high frontend geometry, demonstrate the meaningfulness of this approach. Figure 40 shows high bumper vehicles with high BLE-RL, but still located forward of WAD Page 54 out of 79

55 Figure 40: High vehicles with BLE < WAD For these examples, the BLE-RL WAD is located beyond 954 mm, thus the impact height of the FlexPLI-UBM rubber is to be adjusted to a value between 26 mm and 72 mm above ground level, depending on the exact WAD of the BLE-RL WAD. Figure 41 gives examples of vehicles with extraordinary high BLE-RL, with their WAD beyond 1000 mm. Figure 41: High vehicles with BLE > WAD In these occasions, the FlexPLI-UBM rubber is always to be tested with its lower edge 72 mm above ground level. Page 55 out of 79

56 4.3.3 Derivation of Impactor Thresholds Comparative simulations with the FlexPLI-UBM rubber and the human body model THUMSv4, as reported in SENIORS Deliverable 3.2b, resulted in good transition equations for the femur bending moment during simulations without unintended interaction between the upper body mass and the vehicle front. In Figure 42 the maximum peak femur bending moments, peak tibia bending moments and MCL elongations during THUMSv4 simulations are plotted against the results with FlexPLI Baseline, FlexPLI-UBM rigid and FlexPLI-UBM rubber for all impacts against the Sedan vehicle category (Compact Car, Limousine and SAE Buck three times seven data points altogether). Regarding the elongation of the medial collateral ligament MCL and the bending moment of the tibia, no satisfactory correlation could be established Femur max Bending Moment [Nm] [mm] y = 0,4834x + 173,66 R² = 0,7657 R² = 0, R² = 0, Tibia max Bending Moment [mm] [Nm] y = 0,6439x + 104,47 R² = 0, R² = 0,3258 R² = 0, MCL Elongation [mm] y = 0,9042x - 0,5234 R² = 0,1801 R² = 0,213 R² = 0,0192 Figure 42: Correlation of maximum femur and tibia peak bending moments and MCL elongation (THUMSv4 vs. FlexPLI) Sedan category. Thus, a further investigation focused on maxima correlation between FlexPLI Baseline and FlexPLI-UBM rubber. Here it could be found that the tibia and MCL results correlated well with coefficients of determination of 0.61 and 0.57 while a poor correlation could be found for the femur sections, only, see Figure 43. Page 56 out of 79

57 Femur max Bending Moment [Nm]: FlexPLI vs FlexPLI-UBM rubber Tibia max Bending Moment [Nm]: FlexPLI vs FlexPLI-UBM rubber Max. MCL Elongation [mm]: FlexPLI vs FlexPLI-UBM rubber Figure 43: Correlation of maximum femur and tibia peak bending moments and MCL elongation (FlexPLI Baseline vs. FlexPLI-UBM rubber ). Since the test with the FlexPLI-UBM is meant as a unique replacement of the lower legform to bumper as well as the upper legform to WAD 775 test, the FlexPLI- UBM rubber impactor thresholds are to be derived by a transformtion of human femur injury criteria and FlexPLI Baseline tibia and MCL limits Femur Injury risk functions for the thigh and the bare femur loaded at mid shaft and distal third in three point bending tests were developed by Kerrigan et al. (2004). Figure 44: Injury risk curves for femur bending moment (Kerrigan et al., 2004). These injury risk functions were based on test data that was normalised to represent the 50 th percentile (average height) pedestrian. The normalisation was based on the Page 57 out of 79

58 geometry of a single, commercially available anatomical femur model that was described as 50 th percentile. However, review of the data suggest that the mean height of the subjects was already close to 50 th percentile: The mean height of subjects in the Kerrigan data was m, with 29 out of 24 subjects being male and five female. The mean height for males and females in the UK is m and m 1, which suggests that the mean height for the Kerrigan proportion of males and females would be m (or 1.7% less than the mean height of subjects in the Kerrigan data). A similar height is also defined for 50th percentile male ATDs. Logically, therefore, the normalisation should have had very little effect on the mean bending moment for fracture. However, the Kerrigan normalisation reduced the mean bending moment for fracture from 465 Nm to 386 Nm, a reduction of 17% - which is far greater than the 1.7% difference in height between the Kerrigan sample and the 50th percentile. Therefore, it was decided that the Kerrigan data should be reanalysed to remove the normalisation. The analysis was performed using R-scripts developed in the SENIORS project, building on the survival analysis method published in ISO TS 18506:2014 and adding the capability to use multiple covariates. The analysis indicated that subject age, mass, impact location and presence of flesh (cf. bare femur bone) were all significant covariates. The Kerrigan data contained 34 data points and ideally more than 40 data points would be used for four covariates (for exact data, such as that available in the Kerrigan data set). This would give more confidence in the parameterisation, particularly of the age parameter because there is not much variation in the ages in the dataset. Figure 45 shows the resulting injury risk function for mid-shaft femur fracture, based on a Weibull distribution, for a 65-year-old with a mass of 75 kg. Table 11 shows the mid-shaft femur bending moment associated with different risks of femur fracture for the same data. 1 AdultData: The handbook of adult anthropometric and strength measurements. Department of Trade and Industry, 2002 Page 58 out of 79

59 1 0,9 0,8 0,7 0,6 0,5 0,4 0,3 0,2 0,1 0 Kerrigan (2004): 65 yo; 75 kg; mid-shaft; with flesh Figure 45: Injury risk curve for femur bending moment, based on analysis of data from Kerrigan et al., Page 59 out of 79

60 Table 11: Femur fracture risk for 65-year-old, 75 kg thigh loaded at the mid-shaft, based on analysis of data from Kerrigan et al. (2004). Femur fracture risk (%) Mid-shaft bending moment (Nm) As reported in SENIORS Deliverable 2.5b, the FlexPLI represents the pedestrian s lower extremities including a simulation of the human flesh. Therefore, the relevant thigh fracture risks at mid shaft derived from the Kerrigan et al. data of 394 Nm (for the 20% risk) and 459 Nm (for the 50% risk) respectively should be taken into account. Finally, the application of the transition equation in Figure 42 results in upper and lower performance limits for the FlexPLI-UBM rubber femur bending moments as follows: Femur max, UPL (FlexPLI-UBM rubber ) = (0,4834* ,66)Nm = 364Nm Femur max, LPL (FlexPLI-UBM rubber ) = (0,4834* ,66)Nm = 396Nm For the development of a VRU rating scheme, the relevance of VRU femur injuries related to knee (MCL) and tibia injuries is estimated at 30%. Thus, when keeping the previous rating scheme of Euro NCAP with a maximum of 1 point per grid point, a maximum of 0.3 points can be achieved per impact point in case of a maximum performance, i.e. meeting the upper performance limit. In case of exceeding the lower performance limit, no femur points are awarded to the test point. Between the upper and the lower performance limit, a sliding scale is applied, as illustrated in Figure 46. Page 60 out of 79

61 Figure 46: Sliding Scale for the femur assessment of the FlexPLI-UBM rubber Tibia During the development and technical evaluation of the FlexPLI, two injury risk curves for the risk of tibia fracture were discussed (Konosu et al., 2009). The first one developed by Takahashi (2009) was related to the risk of the 50 th percentile male to suffer tibia fracture took into account scaled male and female PMHS data from Nyquist et al. (1985) and Kerrigan et al. (2004) under modification of the standard tibia length and standard tibia plateau height, and making the assumption that the height scale factor and length scale factor should correlate to each other. The Weibull survival model was used to develop the injury probability function. The second risk curve developed by Pastor and Zander (2010-2) was taking into account scaled male PMHS data from Nyquist et al. and using the standard tibia plateau height provided by DIN German anthropometric database (2005). The cumulative Gaussian distribution was used to develop the injury probability function. Scaled Fracture Moment *1) *1): according to formula M scaled =[(L ref /L)³]*M max under consideration of DIN standardized tibia heights Source: Pastor C. (2009) Figure 47: Underlying data for FlexPLI: injury risk curves for tibia bending moment. Page 61 out of 79

62 As upper performance limit a 20% risk for human tibia bone fracture was chosen by Euro NCAP, calculating the average value for both risk curves (300 Nm and 246 Nm respectively) to 273 Nm. According to the FlexPLI vs. HBM transition equation the FlexPLI threshold was determined as follows: Risk curve I: Tibia max (FlexPLI) I = * Tibia max (HBM) = 305 Nm Risk curve II: (consideration of 10% scatter): Tibia max (FlexPLI) II = (1.259 * Tibia max (HBM) )*1.1 = 258 Nm Tibia max, UPL (FlexPLI)= (Tibia max (FlexPLI) I + Tibia max (FlexPLI) II ):2 = 282 Nm For the lower performance limit, the requirement proposed for type approval (UNECE, 2015) was chosen. Here, the 30% human bone fracture risk at 330 Nm human tibia bending moment according to the risk curve developed by Takahashi was directly transferred into the corresponding FlexPLI threshold: Tibia max (FlexPLI) LPL = * Tibia max (HBM) = 343 Nm This value was rounded to 340 Nm. Using the tibia transition equation from Figure 43 and the current Euro NCAP upper and lower performance limits of 282 and 340 Nm results in upper and lower performance limits for the FlexPLI-UBM rubber tibia bending moments as follows: Tibia max, UPL (FlexPLI-UBM rubber ) = (0.8279* )Nm = 287 Nm Tibia max, LPL (FlexPLI-UBM rubber ) = (0.8279* )Nm = 335 Nm For the development of a VRU rating scheme, the relevance of VRU tibia injuries related to knee (MCL) and femur injuries is estimated at 30%. When keeping the previous rating scheme of Euro NCAP with a maximum of 1 point per grid point, a maximum of 0.3 points from tibia assessment can be achieved per impact point in case of a maximum performance, i.e. meeting the upper performance limit. When exceeding the lower performance limit, no tibia points are awarded to the test point. Page 62 out of 79

63 Between the upper and the lower performance limit, a sliding scale is applied, as shown in Figure 48. Figure 48: Sliding Scale for the tibia assessment of the FlexPLI-UBM rubber Knee For a calculation of the risk of injuries to the human knee, Takahashi (2009) averaged the 50% risk for knee injury derived from risk functions developed by Ivarsson et al. (2004) and Konosu et al. (2004) (see Figure 49), resulting in a knee bending angle (KBA) of % AIS2+ injury risk (18,2 ) Figure 49: Injury risk curves for human knee bending angle according to Konosu (2004, left) and Ivarsson et al. (2004, right). Page 63 out of 79

64 This value was transferred to a corresponding elongation of the medial collateral ligament (MCL-EL) under application of the transition equation: MCL-EL human model = 0.835*KBA human model and under the assumption KBA human =KBA human model to a 50% risk for MCL rupture as MCL-EL human model = 0.835*19 = 15.87mm. According to the FlexPLI vs. HBM transition equation and incorporation of the effect of muscle tone (Lloyd et al., 2001) the FlexPLI threshold was calculated as follows: MCL-EL max (FlexPLI) = (0.5584*15.87 mm mm)*1.1 = 21.1 mm. Under comparison with the legal requirement of 19 for the EEVC WG 17 legform impactor according to Regulation (EC) No. 78/2009 and the 00 series of UN-R 127, it was found that the calculated MCL requirement, correlating to a EEVC WG 17 legform impactor knee bending angle of approx. 17, would have a higher stringency. Therefore, and since this value was also in line with a correlation study between FlexPLI MCL elongation and EEVC WG 17 legform impactor knee bending angle performed by BASt (Zander O., 2010), the FlexPLI threshold in UN-R was set to 22 mm: MCL-EL max(gtr) (FlexPLI) = MCL-EL max (FlexPLI) * 1.05 = 22 mm Euro NCAP used the 22 mm knee elongation as its lower performance limit. For the upper performance limit, an MCL elongation correlating to 15 knee bending angle of the EEVC WG 17 legform impactor - as the previous Euro NCAP upper performance limit - was chosen according to the abovementioned correlation study: Page 64 out of 79

65 MCL-EL max, UPL (FlexPLI) = 0.91*KBA+5.37 = 0.91* = 19 mm Using the MCL transition equation from Figure 43 and the current Euro NCAP upper and lower performance limits of 19 and 22 mm results in upper and lower performance limits for the FlexPLI-UBM rubber MCL elongations according to the following equations: MCL-EL max, UPL (FlexPLI-UBM rubber ) = (0.9487*19mm mm) = 25mm MCL-EL max, LPL (FlexPLI-UBM rubber ) = (0.9487*22mm mm) = 28mm For the development of a VRU rating scheme, the relevance of VRU knee (MCL) injuries related to femur and tibia injuries is estimated at 40%. Here it needs to be added, that ACL/PCL ruptures in lateral vehicle to pedestrian accidents are expected to be covered by the protection of the medial collateral ligament (compare Figure 50); thus, a separate injury criteria seems not necessary. Figure 50: Knee injury mechanisms for ACL, PCL and MCL ruptures (Teresiński et al., 2001). When keeping the previous rating scheme of Euro NCAP with a maximum of 1 point per grid point, a maximum of 0.4 points from the assessment of MCL can be achieved per impact point in case of a maximum performance, i.e. meeting the upper performance limit. When exceeding the lower performance limit, no MCL points are awarded to the test point. Between the upper and the lower performance limit, a sliding scale is applied, as shown in Figure 51. Page 65 out of 79