Obstacle Avoidance Path Planning for Robotic Arms Using a Multi-Strategy Collaborative Bidirectional RRT* Algorithm

Area of Science:

• Robotics and Control Systems
• Artificial Intelligence and Machine Learning
• Computational Geometry

Background:

• Traditional RRT* algorithm faces challenges including sampling bias, slow convergence, inefficient path searching, and poor path smoothness.
• These limitations hinder effective path planning in complex environments.
• Addressing these issues is crucial for advancing autonomous systems.

Purpose of the Study:

• To propose an improved RRT* algorithm that overcomes the limitations of the traditional approach.
• To enhance path planning efficiency, convergence speed, and path quality (smoothness, length, safety).
• To provide a superior path planning solution for 3D environments.

Main Methods:

• Implemented a dynamic ellipsoidal sampling strategy for adaptive exploration of the path space.
• Utilized a bidirectional RRT* algorithm with dynamic goal-biased sampling and heuristic search for accelerated convergence.
• Integrated an improved artificial potential field method for directional guidance and employed path pruning with cubic B-spline interpolation for path smoothing.

Main Results:

• The improved RRT* algorithm demonstrated significant reductions in planning time (58-90%) and path nodes (31-91%) compared to existing algorithms.
• Achieved an average path length reduction of 8-20% across simple and complex 3D environments.
• The final generated trajectory exhibited continuous curvature, suitable for practical tracking applications.

Conclusions:

• The proposed enhanced RRT* algorithm offers substantial improvements in path planning efficiency and quality.
• The combination of dynamic sampling, bidirectional search, goal-biasing, and smoothing techniques results in superior performance.
• This algorithm provides a robust and efficient solution for complex 3D path planning tasks.