Dynamic path planning of an omni-directional robot in a dynamic environment

by Wu, Jianhua.

The objective of this research is to establish a method of local motion planning for mobile robots which could be applied to a dynamic environment, wherein the obstacles are moving and their velocity profiles are not pre-known. This is accomplished by developing a hybrid system in which a global deliberate approach is applied to determine the motion in the desired path line (DPL), and a local reactive approach is used for moving obstacle avoidance. The dissertation shows through theoretical analysis, simulation, and experiment that the developed method can be effectively used to avoid collisions with obstacles in a dynamic environment. A unique modeling of a three-wheeled omni-directional robot has been developed. The equations of motion (EOM) are coupled nonlinear equations. The novel concepts of velocity and acceleration cones are proposed for determining the kinematic and dynamic constraints of a three-wheeled omni-directional robot. The velocity cone and acceleration cone are the feasible velocity and acceleration ranges in the Cartesian coordinates which are suitable for any orientation of the robot. A hybrid approach has been developed for local motion planning. A global deliberate approach has been applied to the motion along the desired path line (DPL) while a local reactive approach is used to avoid collisions with obstacles. For the experiments, a video camera is used as the sensor to detect the unknown motion of the obstacles. The kinematic and dynamic constraints of the robot have been taken into account in the developed method. Approved: Robert L. Williams II Professor of Mechanical Engineering vi