An Efficient Online Trajectory Generation Method Based on Kinodynamic Path Search and Trajectory Optimization for...

Area of Science:

• Robotics
• Artificial Intelligence
• Control Systems

Background:

• Human-robot collaboration is increasing in shared workspaces, necessitating robust safety measures.
• Traditional motion planning struggles with dynamic environments and uncertain obstacle movements.
• Ensuring safety in human-robot coexistence is critical as robots operate outside fixed perimeters.

Purpose of the Study:

• To develop an efficient online trajectory generation method for manipulator autonomous planning in dynamic environments.
• To enhance safety and efficiency in human-robot interaction within shared workspaces.
• To address the limitations of traditional motion planning in uncertain, dynamic settings.

Main Methods:

• An efficient kinodynamic path search algorithm considering link constraints for initial trajectory generation.
• B-spline convex hull property for trajectory optimization, minimizing collision cost, improving smoothness, and ensuring dynamic feasibility.
• Constraint-relaxed link collision avoidance using quadratic programming to prevent link-obstacle and self-collisions.

Main Results:

• The proposed method generates smoother, collision-free trajectories.
• Achieves higher success rates and reduces planning time compared to state-of-the-art methods.
• Demonstrated effectiveness through detailed simulation and real-world experiments.

Conclusions:

• The developed online trajectory generation method significantly improves manipulator planning in dynamic environments.
• It enhances safety and efficiency for human-robot collaboration.
• The approach offers a viable solution for real-world applications requiring reliable robot navigation.