A preliminary experimental study

Transcription

1 Jean-Christophe FAUROUX Guillaume DOUARRE LaMI / IFMA / UBP Clermont-Ferrand, France Laboratoire de Mécanique & Ingénieries skid-steering on a 6x6 all-terrain vehicle A preliminary experimental study Philippe VASLIN KOKOON project Philippe.Vaslin@isima.fr LIMOS / UBP Clermont-Ferrand, France IFToMM 27, The 2th World Congress in Mechanism and Machine Science FRANCE IFToMM 27, The 2th World Congress in Mechanism and Machine Science June 7 2, Besançon,

2 to the problem Two existing principles for steering a wheeled vehicle Ackermann steering Our purpose To keep the robustness and mechanical simplicity of skid-steering vehicles Trying to minimize the energy consumption of such vehicles during steering A complex phenomenon, difficult to model Preliminary experimental results to check consistency Evaluate the ripping forces Try to minimize them......by adjusting normal contact forces IFToMM 27, The 2th World Congress in Mechanism and Machine Science 2

3 Ackermann steering Ackermann Ackermann (87) / Jeantaud (87) For wheeled vehicles that steer with minimal skid With N axles, (N-) axles must steer at least Mostly for hard ground Tricycle scooter, Piaggio MP3 C Mobile crane, Liebherr steering axles out of 8 Power lift truck, Manitou M26 Mobile mortar, Patria NEMO Minimal skid vs. complexity IFToMM 27, The 2th World Congress in Mechanism and Machine Science Semi trailer steering axle out of 5 + R 3

4 Ackermann For wheeled and tracked vehicles Mechanical simplicity (no steering mechanism, only suspension) Wheels cannot roll laterally Mostly for low friction / soft grounds Two versions of the same mobile mortar Ackermann and Skid steering tracked, Patria AMOS C Amphibian ATV, Oasis LLC Max 6x6 Tracked tractor, Caterpillar D6R III Mobile robot, Pioneer 3 AT Simplicity vs. turning resistance IFToMM 27, The 2th World Congress in Mechanism and Machine Science AT wheelchair, Modul Evasion 4

5 Overview of the 6x6 Kokoon vehicle An ATW (All-Terrain Wheelchair) 6x6 transmission and 2x33W electric engines Oleo-pneumatic suspension on all wheels Climbing capacity: 2cm obstacles, 36% slopes Speed: 8 km/h Autonomy: 4h Overview Transmission Suspension IFToMM 27, The 2th World Congress in Mechanism and Machine Science Designed at IFMA from 999 5

6 Overview of the 6x6 Kokoon vehicle Multiple functions Sliding sport seat Top case or transporting platform Overview Transmission Suspension Safety harness Improved access: sliding seat Controlled by joystick Roll bar Lead batteries 24V 6Ah Transport position 2 DC electric engines of 33W each 6x6 all terrain vehicle Independent oleo pneumatic suspensions Skid steering for robustness and all terrain mobility A reconfigurable vehicle A research platform with sensors IFToMM 27, The 2th World Congress in Mechanism and Machine Science Research configuration with onboard camera, double acquisition laptop and suspension sensors 6

7 transmission Belt transmission: 3 wheels at the same speed Compatible with suspension displacements Good efficiency Clutching Overview Transmission Suspension IFToMM 27, The 2th World Congress in Mechanism and Machine Science 7

8 suspension Independent wheels 3 cm cm Length = 75 cm Width = 3 cm 7 cm Track width = 93 cm Wheel base = 47 cm 4 cm Overview Transmission Suspension cm Trailing arm suspensions Wheel diam. = 4 cm Wheel width = 7 cm 47 cm cm Robust Easy to adjust Oleo-pneumatic shock-absorber IFToMM 27, The 2th World Congress in Mechanism and Machine Science 8

9 The force plate Steady state circular turning A force plate to measure contact reaction forces Rx Ry Rz longitudinal force provided by the engines lateral force due to ripping normal force due to mass dispatching Z Cir cu Force plate Mass Suspensions Turning lar t X raje cto ry Gro und For ce pla Y te IFToMM 27, The 2th World Congress in Mechanism and Machine Science 9

10 The force plate 6cm A six component force plate (TSR, France) Force measurement ranges: Force plate Mass Suspensions Turning Rx: N Ry: 9N 8cm Rz: 2N 3 two component strain gauge transducers at 2 3 tangential components (Rx Ry) 3 normal components (Rz) Acquisition chain: 6 channels, 6 bits, Hz, Labview Wrench = 6x6 Calibration matrix x Tensions Fx Fy Fz =[ C ] Mx My Mz V V2 V3 V4 V5 V6 Wooden box Z Wooden box Force transducers Ground level Top plate Base plate X IFToMM 27, The 2th World Congress in Mechanism and Machine Science Y

11 Mass dispatching Centre of mass is computed with the CAD model Hypothesis : uniform density in each part Empty mass: 367 kg Force plate Mass Suspensions Turning IFToMM 27, The 2th World Congress in Mechanism and Machine Science

12 Mass dispatching al measurement 6 scales with the same height Force plate Mass Suspensions Turning CAD CM Central axle axis Real CM Lifting rectangle Good correlation / CAD IFToMM 27, The 2th World Congress in Mechanism and Machine Science 2

13 Suspension reconfiguration Reconfigurable suspension Reference configuration Three identical suspensions Includes a 83kg driver 8 kg Force plate Mass Suspensions Turning 83 kg Modified configuration Differentiated suspensions Central suspension becomes more loaded 8 kg ( 26%) A new mass dispatching 25 kg (+58%) z Reference dimensions 58 kg Adjustment with hs Initial hs = 45mm New hs = 45mm x Parameter Value vs 8 mm hs std 45 mm hs mod 45 mm ls 2 28 mm l2 7 mm la 35 mm e 5 mm IFToMM 27, The 2th World Congress in Mechanism and Machine Science 2 kg ( 35%) hs O S θs vs Fs ls B θa C l2 Fw e la S2 3

14 Turning experiment Three types of trajectories for the external wheels Radius R Infinite ( => straight line) 6m 3m Acquisition at Hz is enough for low speed (around 4 5 kph) Duration: from.5s (straight line) to 2.5s (3m turn) Force plate Mass Suspensions Z Y R= 3m R Turning X Li ne s dr aw n wi th = 6m flo ur IFToMM 27, The 2th World Congress in Mechanism and Machine Science 4

15 Typical curves Curves of Rx, Ry, Rz are provided One or two wheels on the force plate at the same time Five zones appear on the time axis Average values on each zone (vibrations on gravel and electric perturbations) Rz = 627N Rz2 = 972N Rz3 = 95N (one side only) Main results: Wheels 2 and 3 prevail for Rz > confirms weight dispatching 2 Reaction forces (N) Wheel Standard suspension Wheels and 3 prevail for Ry and turning moment Wheel 2 generates most of the traction force Rx (belt sliding?) Wheels +2 Wheel 2 Wheels 2+3 Wheel 3 Rx (N) Ry (N) Turn R = 6m Rz (N) 5 Curves Reference Modified COP Explanation 5 5,5 Time (s),5 IFToMM 27, The 2th World Congress in Mechanism and Machine Science 2 5

16 2 Wheels +2 Reaction forces (N) Standard suspension Straight line Wheels 2+3 for reference vehicle Rx (N) Ry (N) Rz (N) 5 Wheel Wheel 2 Wheel 3 Rz Influence of radius R Rz does not changes with R 5 Rx Ry Ry is quasi null in straight line Ry increases when R decreases 5,5 Time (s),5 2 Reaction forces (N) Standard suspension Turn R = 6m 5 Rx is quasi null in straight line Rx increases when R decreases This phenomenon was experienced by the driver Rz 5 Rx Curves Reference Ry 5,5 Time (s),5 2 2 Reaction forces (N) Standard suspension Turn R = 3m Modified 5 COP Explanation Rz 5 Rx Ry IFToMM 27, The 2th World Congress in Mechanism and Machine Science 5,5 Time (s),5 2 2,5 6

17 2 Reaction forces (N) Standard suspension Straight line Wheels +2 2 Wheels 2+3 for modified vehicle Reaction forces (N) Modified suspension Straight line Rx (N) Ry (N) Rz (N) 5 Wheel Wheel 2 5 Wheel 3 Rz 5 Rx Rx Ry Ry 5 5,5 Time (s),5 2 Reaction forces (N) Standard suspension Turn R = 6m Rz 5 Rx Rx Ry Ry 5 5,5 Time (s),5 2 2 Reaction forces (N) Standard suspension Turn R = 3m Modified Time (s),5 Turn is shorter on modified vehicle Reaction forces (N) Modified suspension Turn R = 3m 5 A 36% decrease of Ry on Wheel 3 COP,5 2 5 Rz 5,5 5 Reference A 38% decrease of Rx on Wheel 2 Time (s) Reaction forces (N) Modified suspension Turn R = 6m 5 Curves,5 2 Rz 5 Explanation Rz Rz 5 5 Rx Rx Ry Ry Science IFToMM 27, The 2th World Congress in Mechanism and Machine 5 5,5 Time (s),5 2 2,5,5 Time (s),5 7

18 Centre of pressure (COP) COP = the point where the moment of contact wrench is null Phases are visible (one or two wheels on the force plate) The COP follows circles => Trajectories were correct 6 X (cm) 5 Curves Reference Modified COP Explanation Going straight 4 Turn R=6m Turn R=3m 3 2 Y (cm) IFToMM 27, The 2th World Congress in Mechanism and Machine Science

19 Physical explanations Strong influence of mass dispatching on turning ability Analogy with the behaviour of a road tire Lateral force Ry increases with normal force Rz Lateral force Ry depends also on slip angle s Linear behaviour up to sl During skid steering, two slip angles s and s3 appear on wheels and 3 Force platform Lateral force RY X s Curves Reference Modified COP Explanation RZ > RZ2 > RZ3 V V2 V3 s3 RZ RZ2 RZ3 Slip angle Y s sl Circular trajectory of radius R IFToMM 27, The 2th World Congress in Mechanism and Machine Science 9

20 can be greatly improved......only by minor adjustments on the suspensions cm adjustments only decreased normal force of 3% on front and rear axles Propulsion forces decreased of 3% when steering Multi-axles vehicles without steering systems are robust. They can now be efficient when skid-steering IFToMM 27, The 2th World Congress in Mechanism and Machine Science 2

21 Replace belts by chains to suppress sliding Add 6 force sensors inside the wheels for continuous measurement Build analytical model Towards an active correction system TSR force sensor in a wheel Acknowledgement ANVAR : French National Agency for Development of Research TIMS Research Federation Mobility pole Michelin Company...and many more on IFToMM 27, The 2th World Congress in Mechanism and Machine Science 3 Videos 2