Multi-layered Safety of Redundant Robot Manipulators via Task-oriented Planning and Control

TL;DR

This work proposes a taskoriented planning and control framework to achieve multi-layered safety while maintaining efficient task execution and demonstrates that the proposed framework can effectively handle uncertain static or dynamic obstacles, and perform disturbance resistance in manipulation tasks when unforeseen contacts occur.

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.

Figures (8)

• Figure 1: When a robot operates in an open workspace, static and dynamic obstacles, as well as unknown contacts, need to be considered to ensure safety, while minimizing their impact on task execution is desirable.
• Figure 2: A schematic diagram depicting the various components of the proposed framework. The high-level planning module is to avoid known obstacles, while the cascaded control module aims to handle unknown contacts.
• Figure 3: Illustration of multi-layered safety in robot manipulation. S1 and S2 are realized in the trajectory planner. S3 is handled by the torque controller.
• Figure 4: Simulations of obstacle avoidance. (a) and (b) are snapshots. (c) shows the distances between the robot links and the obstacle for the top-left scenario. (d) compares solution times using different shooting methods and horizons.
• Figure 5: Case study #1 of S1 and S2 under static obstacles. (a) and (b) are experimental snapshots. (c) plots the end-effector trajectories. (d) presents the closest distances $d_i$ between the link 5-6 and the cabinet's top board.