Collision-free Trajectory Planning for Robotic Agents Based on Metaheuristic Optimization
To ensure the safe motion of robotic agents in complex three-dimensional(3-D)environ-ments,a collision-free trajectory planning scheme based on metaheuristic optimization was proposed.Trajectory nodes were interpolated using quintic splines to achieve sufficient smoothness of the mo-tion trajectory,with zero velocities and accelerations at both the starting and target points.The opti-mization objective was to minimize the trajectory length while constraining the minimum distance be-tween the trajectory and obstacles,formulating it as an optimization problem.This scheme was solved using the metaheuristic-based grey wolf optimizer.Subsequently,two simulations were con-ducted in complex 3-D terrains to evaluate the proposed scheme.The results demonstrated that this scheme effectively identified the predetermined number of optimal trajectory interpolation nodes with-in the planning space,with the trajectory length formed by quintic spline interpolation being mini-mized,and avoiding any collisions with obstacles.
robotic agenttrajectory planningcollision-freequintic spline interpolationgrey wolf optimizermetaheuristic
NETL
NSTL
万方数据
