Co-optimizing Physical Reconfiguration Parameters and Controllers for an Origami-inspired Reconfigurable Manipulator

2025-04-14

AI Summary

Tendon-driven origami robotic arms face significant challenges in jointly optimizing configuration and control under dynamic task or environmental changes due to strong coupling between physical reconfiguration and control. Method: This paper proposes a unified reinforcement learning framework that co-optimizes physical reconfiguration parameters—specifically joint stiffness—and control policies within a proximal policy optimization (PPO) paradigm. It introduces joint stiffness as a learnable physical parameter for the first time in PPO-based co-optimization, abandoning the conventional fixed-configuration assumption. A high-fidelity forward dynamics model is established via the minimum potential energy principle, enabling end-to-end co-learning of stiffness modulation and tendon displacement control. Results: Experiments demonstrate successful execution of complex 3D target-reaching tasks infeasible under fixed-stiffness configurations, with substantial improvements in task adaptability, obstacle avoidance capability, and motion robustness—establishing a novel paradigm for “integrated configuration-control adaptation” in reconfigurable robots.

Technology Category

• Intelligent Robots: Learning & Optimization for ROB
• Search and Optimization: Learning to Search
• Planning, Routing, and Scheduling: Learning for Planning and Scheduling

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Responsible Web: Machine-in-the-loop, human agency and autonomy
• User Modeling, Personalization and Recommendation: Federated recommendation systems and personalization

Abstract

Reconfigurable robots that can change their physical configuration post-fabrication have demonstrate their potential in adapting to different environments or tasks. However, it is challenging to determine how to optimally adjust reconfigurable parameters for a given task, especially when the controller depends on the robot's configuration. In this paper, we address this problem using a tendon-driven reconfigurable manipulator composed of multiple serially connected origami-inspired modules as an example. Under tendon actuation, these modules can achieve different shapes and motions, governed by joint stiffnesses (reconfiguration parameters) and the tendon displacements (control inputs). We leverage recent advances in co-optimization of design and control for robotic system to treat reconfiguration parameters as design variables and optimize them using reinforcement learning techniques. We first establish a forward model based on the minimum potential energy method to predict the shape of the manipulator under tendon actuations. Using the forward model as the environment dynamics, we then co-optimize the control policy (on the tendon displacements) and joint stiffnesses of the modules for goal reaching tasks while ensuring collision avoidance. Through co-optimization, we obtain optimized joint stiffness and the corresponding optimal control policy to enable the manipulator to accomplish the task that would be infeasible with fixed reconfiguration parameters (i.e., fixed joint stiffness). We envision the co-optimization framework can be extended to other reconfigurable robotic systems, enabling them to optimally adapt their configuration and behavior for diverse tasks and environments.

Problem

• Optimizing reconfigurable parameters and controllers for origami-inspired manipulators
• Co-optimizing joint stiffness and tendon displacements for task adaptation
• Developing a framework for reconfigurable robots to adapt to diverse tasks

Innovation

• Co-optimizes joint stiffness and tendon control
• Uses reinforcement learning for parameter optimization
• Predicts shape via minimum potential energy model