Trajectory Planning for a Mobile Robot

Abstract

Robotics, when successfully carried out, is concerned with artifacts that optimize their own behavior in the light of information using sensors and actuators. The information includes both that which is stored and that which is immedi ately available from the sensors. In this paper the problem addressed is how the speed and the steer angle of a path curvature limited robot-vehicle, one that has to move in order to turn, may be altered in order to achieve desirable changes in its position and orientation. A kinematic description of both a tricycle and "turtle" vehicle architecture is developed. A comparison is made between the two architectures. The effect of modulating the steer angle of a path curvature limited vehicle, a tricycle, with a gaussian envelope is illustrated for various amplitudes of the gaussian and speed of the robot-vehicle. The effect of the time concat enation of the steer angle with two or more gaussian enve lopes shows that a range of activities relevant to the functional needs of a robot-vehicle are possible. A number of other "modulating" functions are identified, and it is suggested that these deserve further investigation.

Directed Sonar Sensing for Mobile Robot Navigation

“…More sophisticated vehicle kinematic and dynamic models have been presented in the literature, such as those provided by Muir [107], Steer [122] and Alexander [1]. The point kinematic model has proven adequate in our experiments, and we feel its use is justified because our concern rests with the perception issues of position estimation.…”

Optimal kinematic path tracking control of mobile robots with front steering

“…On the other hand, the lateral displacement of point M is due to change in the vehicle orientation during motion; the lateral motion in XOY coordinates and the orientation change are interrelated according to where y M is the lateral displacement of M (y M : the lateral velocity in the ground attached coordinates). In this sense the lateral velocity of point C is y c = V u A0 + co(MC) (9) where y c is the lateral velocity of point C and co(MC) denotes the magnitude of co x MC. If the position error e d is based on the deviation of the mass centre then the rate of change of the position error and the lateral velocity of point C are the same for a straight path and would be as follows for a curved path:…”

Motion feasibility of a wheeled vehicle with a steering angle limit

“…The reference path for an automatically guided vehicle around factories or offices generally comprises a concatenation of line-circular segments. 34 The path generated by the available algorithms for an autonomous vehicle depends on the assumption of the shapes of the obstacles in the workspace and the autonomous vehicle, and the algorithms. It generally consists of a collection of line-line, line-circular segments.…”