Integrated assembly and motion planning using regrasp graphs.

Area of Science:

• Robotics
• Artificial Intelligence
• Manufacturing Automation

Background:

• Automated assembly requires sophisticated planning for both object sequencing and precise motion control.
• Existing methods often address assembly and motion planning separately, leading to suboptimal solutions.
• The use of a horizontal surface as a fixture is common in many assembly tasks.

Purpose of the Study:

• To develop an integrated system for simultaneous assembly sequence and motion planning.
• To enable robots to recursively find optimal assembly sequences and corresponding motions.
• To utilize a horizontal surface as a supporting fixture for enhanced planning.

Main Methods:

• The system operates at both assembly and motion planning levels.
• Assembly level: Explores all assembly sequence combinations to identify candidate sequences.
• Motion level: Employs recursive search and regrasp graph construction for pick-and-place motion planning.

Main Results:

• The integrated system successfully plans assembly sequences and motions concurrently.
• Demonstrated efficacy through both simulation and real-world experimental validation.
• The system effectively manipulates objects and integrates them into the assembly using planned motions.

Conclusions:

• The proposed integrated assembly and motion planning system offers a unified approach to complex robotic assembly tasks.
• Recursive search and graph-based methods are effective for optimizing assembly sequences and motions.
• The system's ability to plan integratedly enhances the efficiency and robustness of automated assembly.