A Predictable Obstacle Avoidance Model Based on Geometric Configuration of Redundant Manipulators for Motion Planning

Area of Science:

• Robotics
• Control Systems
• Computational Geometry

Background:

• Robotic manipulators operating in dynamic environments face challenges with real-time obstacle avoidance.
• Ensuring safety for both the manipulator and surrounding individuals is critical.

Purpose of the Study:

• To develop a dynamic, whole-body obstacle avoidance strategy for redundant manipulators.
• To accurately model manipulator-obstacle interactions for collision prediction.

Main Methods:

• Proposed a novel 'triangular collision plane' model based on manipulator geometry.
• Developed three cost functions: motion state, head-on collision, and approach time.
• Integrated these costs into inverse kinematics using the gradient projection method.

Main Results:

• Simulations and experiments demonstrated improved manipulator response speed.
• The proposed method significantly enhanced system safety compared to existing techniques.
• Validated the effectiveness of the triangular collision plane model.

Conclusions:

• The triangular collision plane model offers an effective approach for real-time dynamic obstacle avoidance in redundant manipulators.
• This method enhances operational safety and efficiency in complex environments.