Driver assistance systems and outlook into automated driving

Similar documents
Automated Driving - Object Perception at 120 KPH Chris Mansley

Citi's 2016 Car of the Future Symposium

Új technológiák a közlekedésbiztonság jövőjéért

The Imperative to Deploy. Automated Driving. CC MA-Info, 15th December 2016 Dr. Hans-Peter Hübner Kay (CC/EB4) Stepper

The connected vehicle is the better vehicle!

Pushing the limits of automated driving with artificial intelligence and connectivity

C A. Right on track to enhanced driving safety. CAPS - Combined Active & Passive Safety. Robert Bosch GmbH CC/PJ-CAPS: Jochen Pfäffle

EMERGING TRENDS IN AUTOMOTIVE ACTIVE-SAFETY APPLICATIONS

Cars that think and act automated driving for greater road safety

Cars that think and act automated driving for greater road safety

DA to AD systems L3+: An evolutionary approach incorporating disruptive technologies

On the road to automated vehicles Sensors pave the way!

Tips & Technology For Bosch business partners

On the role of AI in autonomous driving: prospects and challenges

China Intelligent Connected Vehicle Technology Roadmap 1

Siemens ADAS. Collision avoidance as the first step towards autonomous driving

Electrics/electronics Technology Workshop Cayenne

MEMS Sensors for automotive safety. Marc OSAJDA, NXP Semiconductors

THE FAST LANE FROM SILICON VALLEY TO MUNICH. UWE HIGGEN, HEAD OF BMW GROUP TECHNOLOGY OFFICE USA.

What is the potential of driver assistance technologies to reduce the number of road accidents?

Deep Learning Will Make Truly Self-Driving Cars a Reality

Intuitive Driving: Are We There Yet? Amine Taleb, Ph.D. February 2014 I 1

Intelligent Drive next LEVEL

Cars that can think and act for greater road safety Samantha Cockfield Road Safety TAC

Our Market and Sales Outlook

BMW GROUP TECHNOLOGY WORKSHOPS AUTOMATED DRIVING-DIGITALIZATION MOBILITY SERVICES. December 2016

Highly Automated Driving: Fiction or Future?

Active Safety Systems in Cars -Many semi-automated safety features are available today in new cars. -Building blocks for automated cars in the future.

ACTIVE SAFETY 3.0. Prof. Kompaß, VP Fahrzeugsicherheit, 14. April 2016

AI Driven Environment Modeling for Autonomous Driving on NVIDIA DRIVE PX2

The path towards Autonomous Driving

LiDAR Teach-In OSRAM Licht AG June 20, 2018 Munich Light is OSRAM

Safe, superior and comfortable driving - Market needs and solutions

CSE 352: Self-Driving Cars. Team 14: Abderrahman Dandoune Billy Kiong Paul Chan Xiqian Chen Samuel Clark

Deutsche Bank AutoTech Day

WHITE PAPER Autonomous Driving A Bird s Eye View

Vehicle Integration of multiple ADAS HMI Concept and Architecture

Automated Driving: Design and Verify Perception Systems

Optical Products in Automotive Active Safety

BlueBox: Complete Autonomous Vehicle Platform Using NXP Silicon at Each ADAS Node EXTERNAL USE

RESEARCH FUNDING KEY TO AUTONOMOUS DRIVING

ZF Advances Key Technologies for Automated Driving

Bitte decken Sie die schraffierte Fläche mit einem Bild ab. Please cover the shaded area with a picture. (24,4 x 7,6 cm)

THE FUTURE OF AUTONOMOUS CARS

AUTONOMOUS VEHICLES & HD MAP CREATION TEACHING A MACHINE HOW TO DRIVE ITSELF

Annual press conference on April 29, 2015 Press images

Cooperative Autonomous Driving and Interaction with Vulnerable Road Users

AUTONOMOUS VEHICLES: PAST, PRESENT, FUTURE. CEM U. SARAYDAR Director, Electrical and Controls Systems Research Lab GM Global Research & Development

Smart Control for Electric/Autonomous Vehicles

Stereo-vision for Active Safety

Safety for Self-driving Cars

EPSRC-JLR Workshop 9th December 2014 TOWARDS AUTONOMY SMART AND CONNECTED CONTROL

Applying Innovations in Advanced Driver- Assistance Systems to Material Handling

Leveraging AI for Self-Driving Cars at GM. Efrat Rosenman, Ph.D. Head of Cognitive Driving Group General Motors Advanced Technical Center, Israel

Annual press conference on April 30, 2014 Press images

elektrobit.com Driver assistance software EB Assist solutions

Le développement technique des véhicules autonomes

END TO END NEEDS FOR AUTONOMOUS VEHICLES NORM MARKS SEPT. 6, 2018

ZF Mitigates Rear-End Collisions with New Electronic Safety Assistant for Trucks

STRATEGIES FOR THE MOBILITY TRANSFORMATION HANNO MIORINI, ROBERT BOSCH GROUP

World premiere at Hannover Messe: ZF s highly automated forklift can see, think and act

Cooperative brake technology

Nanotechology in Automotive Aplications. Infineon Technologies Romania. Dr. Michael Neuhaeuser General Executive Manager & VP

IN SPRINTS TOWARDS AUTONOMOUS DRIVING. BMW GROUP TECHNOLOGY WORKSHOPS. December 2017

Accelerating the Driving Experience: The Semiconductor Point of View

future of mobility DI STEFANIE PYKA, ROBERT BOSCH AG WIEN

Autonomous Driving by Audi. Dr. Miklós Kiss

REGULATORY APPROVAL OF AN AI-BASED AUTONOMOUS VEHICLE. Alex Haag Munich,

A factsheet on the safety technology in Volvo s 90 Series cars

Using Virtualization to Accelerate the Development of ADAS & Automated Driving Functions

CONNECTED AUTOMATION HOW ABOUT SAFETY?

Leading and shaping change Focus on technology

Integrated ADAS HIL System with the Combination of CarMaker and Various ADAS Test Benches. Jinjong Lee, Konrad Yu-Mi Song, Hyundai-Autron

David Pickett [Volvo Car Australia]; [National Road Safety Forum 2

THE FUTURE OF SAFETY IS HERE

NXP S32X AUTOMOTIVE PROCESSING PLATFORM

Vehicle Diagnostics and Service Concepts Success Factors in the After Sales for e-mobility

Active Safety and Cooperative Systems in the Road Infrastructure of the Future

Enabling automated driving with artificial intelligence and connectivity Dr. David Weidenfelder

Autonomous Driving. AT VOLVO CARS Jonas Ekmark Manager Innovations, Volvo Car Group

Trends in Automotive MEMS. Dr. Jan Peter Stadler Senior Vice President Engineering, Robert Bosch GmbH

About Automated Driving Functions

JANUARY ADVANCED DRIVER ASSISTANCE TECHNOLOGY NAMES AAA s recommendation for common naming of advanced safety systems. NewsRoom.AAA.

The Digital Future of Driving Dr. László Palkovics State Secretary for Education

Future mobility in a dynamic environment. Dr. Rolf Bulander

Driver Assistance & Autonomous Driving

A factsheet on Volvo Cars safety technology in the new Volvo S90

2018 Schaeffler Symposium 9/6/2018 Philip A. George Foundations of Disruption Preparing for the Uncertainty of Tomorrow s Personal Mobility Challenge

Copyright 2016 by Innoviz All rights reserved. Innoviz

Chassis Systems Control Driver assistance systems How much support do German drivers want?

The competitiveness of the European automotive software industry

Syllabus: Automated, Connected, and Intelligent Vehicles

Autonomous Driving Technology for Connected Cars

AEB Fitment - Available. Jaguar XE saloon

Vehicle: Risks and Measures. Co-funded by the Horizon 2020 Framework Programme of the European Union

Journal of Emerging Trends in Computing and Information Sciences

BOSCH SMART MOBILITY & 3S STRATEGY CHRISTOPHER LAKE

A Communication-centric Look at Automated Driving

Love. It s what makes a Subaru, a Subaru.

Acustomer calls and says that an ADVANCED DRIVER ASSISTANCE SYSTEMS WHAT YOU SHOULD KNOW ABOUT

Transcription:

Driver assistance systems and outlook into automated driving EAEC-ESFA2015 Bucharest November 2015 1

General presentation of the Bosch Group Some 45,600 1 researchers and developers work at Bosch: at 94 2 locations worldwide, in a unique network. Each working day, Bosch files 18 patents on average. This makes Bosch one of the world s leading companies for patent applications. 1 As of 01.15 2 R&D locations with >50 associates, as of 01.15 2 RBRO/ENG 03.09.2015 Robert Bosch GmbH 2013. All rights reserved, also regarding any disposal, exploitation, reproduction, editing,

General presentation of the Bosch Group Bosch: four business sectors Bosch Group 49 billion euros in sales, 360,000 associates Mobility Powertrain systems Safety systems Comfort systems Industrial Technology Mobile applications Factory automation & machinery Renewable energy Packaging technology Energy and Building Technology Heating, ventilation, Decentralized Security & safety Communications climate control energy management Consumer Goods Power tools, measuring tools Garden tools Household appliances Sensors & Smart home 3 RBRO/ENG 03.09.2015 Robert Bosch GmbH 2013. All rights reserved, also regarding any disposal, exploitation, reproduction, editing,

General presentation of the Bosch Group Business year 2014 Sales by business sector Mobility Solutions Bosch Group 49 billion euros EUR 33.3 bn 6.3% nominal 7.4% adjusted for currency effects EUR 4.6 bn EUR 6.7 bn EUR 4.2 bn 8.9% nominal 9.9% adjusted for exchange-rate effects 68% of total sales Industrial Technology -2% nominal -1% adjusted for exchange-rate effects 14% of total sales Consumer Goods 5% nominal 7% adjusted for exchange-rate effects 9% of total sales Energy and Building Technology 1.7% nominal 2.6% adjusted for exchange-rate effects 9% of total sales 4 RBRO/-RBBG/GM 03.09.2015 Robert Bosch GmbH 2013. All rights reserved, also regarding any disposal, exploitation, reproduction, editing,

General presentation of the Bosch Group Bosch Locations: 5 Bosch Entities: 4 CLUJ BLAJ TIMISOARA Bosch Service Solution BPO BUCHAREST Headquarter Sale Research and Development Manufacturing Mobility Solutions Energy & Building Technology Industrial Technology Consumer Goods 5 RBRO/-RBBG/GM 03.09.2015 Robert Bosch GmbH 2013. All rights reserved, also regarding any disposal, exploitation, reproduction, editing,

Automotive industry trends Electric - new energy to the powertrain! Connected cars active parts in internet! Automated driving! 6

We are busy 7

We are distracted 8

Some people need assistance 9

and sometimes driving is boring 10

Highly automated driving user experience https://www.youtube.com/watch?v=cvnujployhs 11

The automated car will be reality! 12

Automated driving key technologies Surround Sensors are highly robust in all use cases Driver Monitoring for partly automatic functions Perception and Localization leads to unambiguous & comprehensive 360 environment model Online Map Data is precise & up to date every moment Functional Safety guarantees high standard @ reasonable effort Reasoning and Decision Making allows for correct decisions, even in highly dynamic situations & at incomplete information Motion Control works safe, fast & precise in all dimensions Architecture supports safety, performance & cost targets. 13

Possible / dominant sensor set up in the market Near Range Cameras Ultra Sound Sensors Camera Mid Range- / Corner- Radar Mid/Long Range Radar Head Unit, Digital Maps also LIDAR possible Car-to-X, Connected Vehicle Currently only partly suitable for low car segments due to high cost pressure 14

Comfort & Assistance Safety for everyone Driver assistance systems and automated driving Leading functions of Advanced Driver Assistance Systems Blind Spot Detection/ Lane Change Assist/ Cross Traffic Alert Pedestrian Protection Automatic Emergency Braking Lane Departure Warning/ Lane Keeping Support Evasion Assist / Construction Zone Assist Adaptive Cruise Control Road Sign Recognition Maneuver Assistance Intelligent Headlight Control Leading functions cover comfort and safety 15

Portfolio of current driver assistance functions Light & sight at night functions High Beam Control Adaptive Headlight Control Masked Continuous High Beam Night Vision with Pedestrian Detection Road Sign functions Speed Limits Stop Signs Give Way Signs Additional signs (Pictograms) Fusion with digital map general signs (triangular, rectangular) Object / Surface functions Adaptive Cruise Control Emergency braking on cars Emergency braking on pedestrians Maneuver Assistance Distance information & warning Cross Traffic Alert Forward Collision Warning Speed Bump Assist Adaptive Suspension Lane functions Lane Departure Warning Lane Keeping Assist Lane Change Assist / Blind Spot Construction Zone Assist Narrow Passage Assist Traffic Jam Assist Evasion Assist Driver Drowsiness Detection 16

Road safety influence of driver assistance 100% Installation rates / road fatalities in Germany 80% 60% Safe braking and steering (ABS) 40% 20% Skidding avoidance (ESP) 0% 1999 2001 2003 2005 2007 2009 2011 2013 Source: Bosch, DAT, BASt. Based on total vehicle fleet. 1 Figures estimated 2 ACC and lane keeping support only Driver assistance 2 Road fatalities Number of road fatalities reduced by 60% within last 14 years 90% of all car accidents involving injury are caused by human error Introduction of further driver assistance systems will amplify positive trend 1 17

Degree of automation Driver assistance systems and automated driving Solutions roadmap to the automated car Single sensor Sensor-data fusion Sensor-data fusion + map Auto pilot ACC/lane keeping support Only longitudinal or lateral control Integrated cruise assist Partially automated longitudinal and lateral guidance in driving lane Speed range 0-130 kph Highway assist Partly automatic longitudinal and lateral guidance Lane change after driver confirmation Supervision of surrounding traffic (next lane, ahead, behind) Highway pilot Highly automated longitudinal and lateral guidance with lane changing Reliable environment recognition, including in complex driving situations No permanent supervision by driver Door-to-door commuting (e.g. to work) in urban traffic Strictest safety requirements No supervision by driver Series production 18

New cars on road for automated driving 2013 2015 ->Tests in real traffic conditions accelerate the development of new functions 19 CC-DA/ESV 03.09.2015 Robert Bosch GmbH 2013. All rights reserved, also regarding any disposal, exploitation, reproduction, editing,

How to build an automated car! https://www.youtube.com/watch?v=op84zlabbai&feature=youtu.be 20

Stereo vision-based perception 21

Driver assistance systems and automated driving Perception Capabilities of a Mono Video System 2D information from the left camera Traffic signs: Speed Limits. Stop Signs. Give Way Signs. Additional signs (Pictograms). General signs (triangular, rectangular). Lane markings. Vehicle lights / traffic lights. Object classification (machine learning). 22

Objects from classification Positive-Samples Classification methods: decision trees, neural networks, support vector machines, boosting. Test Labeled sequences Real World Performance Negative-Samples 23 Features: grayscale values, histogram of oriented gradients, codebook features, Haar like features Object detection using classification is knowledge and training based Classification is very much depending on a priori knowledge

Objects from motion optical flow & optical flow 24 Object detection from motion (optical flow) is based on existence and clustering of 3D information point cloud depending on motion state of the camera (car). No appearance based trained prior knowledge is used for object detection

Lukas-Kanade optical flow method - 1981 Assumption: constant displacement for all the pixels in window. Minimization of C(du,dv) by doing partial derivatives with respect to du and dv and setting them equal to 0. 25 Reference: An Iterative image Registration Technique with an Application to Stereo Vision, D. Lukas, T. kanade, April 1981 https://www.ri.cmu.edu/pub_files/pub3/lucas_bruce_d_1981_1/lucas_bruce_d_1981_1.pdf CC-DA/ESV 03.09.2015 Robert Bosch GmbH 2013. All rights reserved, also regarding any disposal, exploitation, reproduction, editing,

Stereo: Map of depth far near Urban scenario with stop at traffic light Disparity (object distance) 26

Z... distance Driver assistance systems and automated driving Disparity/depth estimation for stereo vision Z = f d B Stereo-Video: Measurement principle left camera right camera f... focal length Left and right imager f disparity... d B... base width 27 CC-DA/PJ-VP 03.09.2015 Robert Bosch GmbH 2013. All rights reserved, also regarding any disposal, exploitation, reproduction, editing,

3D visualization of point cloud from disparity Rectified left image Disparity image text 3D point cloud computed from disparity map 28 Department 03.09.2015 Robert Bosch GmbH 2013. All rights reserved, also regarding any disposal, exploitation, reproduction, editing,

Perception Capabilities of a Stereo Video System complete 3D environment information path topography (slopes, bumps, potholes, lane grooves) height obstructions (bridges, trees) path width (narrow passages, parking) complete 3D object information cars, trucks, cycles, pedestrians, guard rails, poles, walls, trees, animals, Stereo Camera is the only sensor which is able to provide comprehensive 2D/3D information 29

Stereo: elevated objects detection high flat Highway scenario with narrow passage Elevation (object height) 30

Surround sensing vehicle sensor concept - Long-range radar - Mid-range radar - 3 rd sensor principle - Stereo-video - Long-range radar - Mid-range radar - Near-range cameras - Ultrasonic sensors (not to scale) 360 surround sensing by combination of different sensors Long- and mid-range radar prerequisite for driving at higher speed Satisfy reliability requirements by using multiple sensors for each area 31

Research and Development in Cluj Current projects Photovoltaic inverters Embedded Software, Development and Testing, Power Management, Online control & monitoring automotive technology for green energy solutions Electric Power Steering Embedded Software development for electronic steering control unit comfort system fulfilling highest safety requirements stereo camera height map on the way to autonomous driving Engineering Service Center Cluj to internal and external customers PCB layout, mechanical design, Driver Assistance Stereo Image Processing Embedded Software Parking & Maneuvering Reliability Engineering Testing and Analysis of electronic devices durable and reliable products competitive services for Bosch and non-bosch customers Analytic Service Center Data analysis, algorithm & sensor development incl. support for AE/PJ-I4.0 virtual world becomes the leading system ultrasonic sensors Competences Project Management System & Application Algorithms Development System Testing Mechanical Design Hardware Embedded Software 2 nd & 3 rd Level Support PCB Layout Outlook Research & Development Services Cluj plant - Electronics Manufacturing Services Local Romanian customers 32 RBRO/ENG 03.09.2015 Robert Bosch GmbH 2013. All rights reserved, also regarding any disposal, exploitation, reproduction, editing,

Collaboration possibilities Open positions available in the R&D Cluj team. Working students/master/diploma @ Bosch: Develop state of the art algorithmic methods in collaboration with Bosch R&D engineers. PhD @ Bosch: Research and innovation in the areas of driver assistance systems. Perception for highly automated driving. Develop new ideas related to driver assistance, autonomous vehicles and related product. Next generation video-based driver assistance systems. Example topics: precise localization, surface modeling, situation interpretation & path planning. 33 More details: http://www.bosch-career.ro

Thank you for your attention! 34