Trajectory Redundancy in Point-to-Point Tracking and Applications to Obstacle Avoidance

Abstract

For tasks which require a robot to track some particular points along a trajectory (instead of the whole trajectory), there exits redundancy. This redundancy, which results from the robot tracking more points than that required for the task, is called trajectory redundancy. This redundancy results in a increase in the feasibility of the task, enabling the possibility of the robot to obtain better performance by satisfying secondary objectives whilst performing the primary objective of tracking the target points.
Point-to-point learning control (LC) has been shown to be an effect tool to deal with the trajectory redundancy since it has the ability to fully explore the increased feasibility resulting
from such redundancy. This is done using the decomposition technique, which is widely used in solving kinematic redundancy, where the primary and secondary tasks are achieved
independently using the appropriate LC algorithms. In order to apply the proposed point-to-point algorithms for tasks such as tracking a target point whilst avoiding an obstacle, the
algorithms are modified and implemented in an online fashion. Simulation results shows good performance from the proposed point-to-point LC online algorithms.

More information

Identifiers

Catalogue record

Export record