Towards Optimized Parallel Robots for Human-Robot Collaboration by Combined Structural and Dimensional Synthesis

Overview

• Parallel robots (PR) have potential for human-robot collaboration (HRC) due to their lower moving masses and higher speeds.
• However, the parallel leg chains increase the risks of collision and clamping.
• This work describes these hazards using kinematics and kinetostatics models to minimize them as objective functions through a combined structural and dimensional synthesis in a particle-swarm optimization.
• Back-drivability is quantified to theoretically guarantee detectability via motor current.
• The largest eigenvalue of the mass matrix formulated in the operational-space coordinates is used as an HRC-relevant objective function to consider collision effects.
• Multi-objective optimization leads to different Pareto-optimal PR structures.

Plain English Explanation

Parallel robots are a type of robot that have multiple arms or "legs" working together. They can move quickly and have lower overall weight compared to other robots, which makes them well-suited for human-robot collaboration. However, the way the legs are connected also increases the chances of the robot colliding with or trapping a human worker.

This research project looked at ways to address these safety concerns. The researchers used mathematical models of the robot's movement and forces to identify the key risks, such as the robot parts getting stuck together or hitting a person. They then incorporated these risk factors into the robot's design optimization process, using a technique called particle swarm optimization to find the best overall robot structure.

An important part of making these robots safe for human interaction is ensuring they can be easily detected and stopped if needed. The researchers quantified a concept called "back-drivability," which means the robot can be easily moved by an outside force like a person pushing on it. This back-drivability helps guarantee the robot's motion can be sensed by monitoring the motor currents.

The optimization also considered how the robot's mass and inertia properties affect the risk of collisions, using a metric called the largest eigenvalue of the mass matrix. Optimizing for this helps reduce the potential for harmful impacts if the robot does collide with a person.

Overall, this research shows how the design of parallel robots can be optimized to improve their safety and suitability for working alongside human employees, through a combination of mathematical modeling and multi-objective optimization techniques.

Technical Explanation

This work explores the use of parallel robots (PR) for human-robot collaboration (HRC) applications. PRs offer advantages like lower moving masses and higher speeds compared to serial robots, but their parallel leg chains also increase the risks of collision and clamping.

The researchers developed kinematics and kinetostatics models to describe these hazards, and incorporated them as objective functions in a combined structural and dimensional synthesis using particle swarm optimization. In addition to the risks within and between the kinematic chains, the team also quantified the back-drivability of the system to theoretically guarantee detectability via motor current.

Another HRC-relevant objective function used was the largest eigenvalue of the mass matrix formulated in the operational-space coordinates, which captures the effects of collision on the robot. The multi-objective optimization process resulted in different Pareto-optimal PR structures.

The results show that this optimization approach led to significant improvements in the HRC-relevant criteria. The researchers found that a Hexa structure (6-RUS) was the most favorable option, considering the objective functions as well as the simpler joint structure.

Critical Analysis

The research provides a comprehensive framework for optimizing parallel robot designs to enhance safety and suitability for human-robot collaboration. By incorporating specific HRC-relevant objectives, such as collision risk and back-drivability, the optimization process goes beyond traditional mechanical design considerations.

However, the work is limited to theoretical modeling and simulation. Real-world implementation and validation would be necessary to fully assess the practical benefits and limitations of this approach. Factors like sensor reliability, control system responsiveness, and human behavioral factors could also influence the actual performance and safety of such collaborative robots in industrial settings.

Additionally, the study focuses on a specific parallel robot architecture (Hexa 6-RUS). Further research could explore the optimization of other parallel robot configurations, or investigate hybrid designs that combine parallel and serial elements to leverage the strengths of both.

Future work could also consider additional HRC-relevant criteria, such as intuitive programming interfaces, human-friendly motion planning, and strategies for seamless human-robot interaction and task coordination.

Conclusion

This research demonstrates a systematic approach to optimizing parallel robot designs for improved safety and suitability in human-robot collaboration scenarios. By modeling and minimizing key hazards like collision and clamping risks, as well as considering back-drivability and mass matrix properties, the optimization process yields Pareto-optimal designs that can enhance the viability of parallel robots for industrial applications involving close human-robot interaction.

The findings suggest that the Hexa 6-RUS parallel robot structure is a promising candidate for HRC, offering a balance of favorable performance characteristics and simpler joint design. While further validation is needed, this work provides a valuable framework for developing safer and more collaborative robotic systems to support the integration of humans and machines in industrial environments.