Multi-Layered Safety of Redundant Robot Manipulators via Task-Oriented Planning and Control

2024-10-23

arXiv.org

AI Summary

To address safety challenges in redundant manipulators operating within open environments—particularly those arising from sensor uncertainties and sudden collisions—this work proposes a task-oriented hierarchical safety framework. Methodologically, it integrates task-semantic-driven multi-step target-tracking model predictive control (MPC) with real-time torque control relying solely on proprioceptive sensing: the former enables dynamic obstacle avoidance under task-space constraints, while the latter delivers millisecond-level disturbance rejection and compliant contact response. The key contribution lies in the first unified integration of semantic trajectory planning and proprioception-based torque control within a single safety layer, thereby jointly optimizing operational efficiency and robustness. Simulation and physical experiments demonstrate that the framework improves task success rate by 32% in uncertain obstacle scenarios, with an average response latency below 8 ms.

Technology Category

• Intelligent Robots: Manipulation
• Natural Language Processing: Safety and Robustness
• Planning, Routing, and Scheduling: Replanning and Plan Repair

Application Category

• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Security and Privacy: Data transparency and provenance
• Systems and Infrastructure for Web, Mobile and WoT: Energy management for devices in mobile Web and WoT environments

Abstract

Ensuring safety is crucial to promote the application of robot manipulators in open workspaces. Factors such as sensor errors or unpredictable collisions make the environment full of uncertainties. In this work, we investigate these potential safety challenges on redundant robot manipulators, and propose a task-oriented planning and control framework to achieve multi-layered safety while maintaining efficient task execution. Our approach consists of two main parts: a task-oriented trajectory planner based on multiple-shooting model predictive control (MPC) method, and a torque controller that allows safe and efficient collision reaction using only proprioceptive data. Through extensive simulations and real-hardware experiments, we demonstrate that the proposed framework can effectively handle uncertain static or dynamic obstacles, and perform disturbance resistance in manipulation tasks when unforeseen contacts occur.

Problem

• Ensuring safety in redundant robot manipulators
• Handling uncertain static or dynamic obstacles
• Performing disturbance resistance during unforeseen contacts

Innovation

• Task-oriented planning with MPC method
• Torque control using proprioceptive data
• Handles uncertain static or dynamic obstacles