Page 449 - Automotive Engineering Powertrain Chassis System and Vehicle Body
P. 449

CHAP TER 1 4. 2       Decisional architecture












               Fig. 14.2-20 Motion and control commands in the parallel parking scenario: (a) motion, (b) steering angle and (c) velocity controls applied.
               the parking place (e.g. the values of T; f max  and v max  obstacles. The basic motion planning problem is readily
               differ for the first and last iterative motion).    illustrated with the concept of configuration space that
                                                                  was introduced in robotics in the late 1970s by Udupa
               14.2.4.4 Experimental run of the                   (1977) and Lozano-Perez and Wesley (1979a). The
               platooning manoeuvre                               configuration of a robot is a set of independent para-
                                                                  meters representing the position and orientation of every
               An experimental run of the platooning SBM in a street is  part of the robot. In its configuration space, a robot is
               shown in Fig. 14.2-21. The linear camera and the infrared  represented as a point, stationary obstacles are repre-
                                                                                            4
               target is shown in Fig. 14.2-22. During the execution of  sented as forbidden regions  and motion planning
               a platooning manoeuvre, the linear camera operates at  between a start and a goal configuration is reduced to
               a frequency of 1000 Hz for providing the relative  finding a path, i.e. a continuous sequence of configura-
               position/orientation parameters of the two vehicles; the  tions, that avoids the forbidden regions.
               accuracy of the measurement has been estimated at    The basic motion planning problem is essentially
               a value of 1 mm for a distance of 10 m. It has experi-  geometric, it deals with collision avoidance of stationary
               mentally been shown that the system is robust according  obstacles and it computes a path, i.e. a geometric curve in
               to various lighting and light reflecting conditions (thanks
               to the camera characteristics, to the pulsing infrared
               target, and to the used filters). Experiments have been
               conductedatspeedsupto60km/h,withdecelerationsupto
                   2
               2m/s . The distance between the vehicles is proportional to
               the speed (see Section 14.2.3.6), with a gap of 0.3 s.

               14.2.5 Motion planning for car-like
               vehicles


               14.2.5.1 Introduction

               The purpose of every robot is to perform actions in its
               workspace (grasping and mating parts, moving around to
               explore or survey, etc.). Carrying out a given action  Fig. 14.2-21 A platoon of two vehicles: a leader Ligier and
               usually implies that a motion be made by the robot hence  a following Cycab.
               the importance, in robotics, of motion planning, i.e. the
               determination of the motion that is to be performed in
               order to achieve a given task. This importance is naturally
               reflected in the number and variety of research works
               that have dealt with motion planning in the past 30 years.
                 Latombe’s (1991) book is undoubtedly the reference
               book for robot motion planning. Its table of contents
               reveals the importance of what Latombe refers to as the
               basic motion planning problem. Six out of ten chapters are
               dedicated to this problem, which is to plan a collision-  Fig. 14.2-22 Experimental setup for platooning: (a) the linear
               free path for a robot moving freely amidst stationary  camera, (b) the first experimental infrared target.


               4
                The set of configurations yielding a collision between the robot and the obstacle.

                    456
   444   445   446   447   448   449   450   451   452   453   454