Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Constrained Passive Interaction Control: Leveraging Passivity and Safety for Robot Manipulators

Zhiquan Zhang, Tianyu Li, Nadia Figueroa

TL;DR

This work addresses safe, passive physical interaction for torque-controlled manipulators in dynamic human environments by constraining a DS-based passive impedance controller with joint-space safety via exponential CBFs. It introduces ECBF formulations for hard joint-limit and self-collision constraints, and soft external-collision and singularity constraints using learnable neural boundary functions (SCA NN, Neural-JSDF) and a neural JSDF model, all integrated into a Relaxed-CBF-QP that prioritizes safety while preserving passivity whenever feasible. The approach is demonstrated in simulation and on a 7-DoF Franka manipulator, achieving constraint satisfaction and passive interaction under perturbations, with real-time performance on standard optimization tools. The framework holds promise for robust, safe manipulation in human-centric settings and highlights avenues to improve efficiency, such as more compact boundary-function representations and broader hardware deployments.

Abstract

Passivity is necessary for robots to fluidly collaborate and interact with humans physically. Nevertheless, due to the unconstrained nature of passivity-based impedance control laws, the robot is vulnerable to infeasible and unsafe configurations upon physical perturbations. In this paper, we propose a novel control architecture that allows a torque-controlled robot to guarantee safety constraints such as kinematic limits, self-collisions, external collisions and singularities and is passive only when feasible. This is achieved by constraining a dynamical system based impedance control law with a relaxed hierarchical control barrier function quadratic program subject to multiple concurrent, possibly contradicting, constraints. Joint space constraints are formulated from efficient data-driven self- and external C^2 collision boundary functions. We theoretically prove constraint satisfaction and show that the robot is passive when feasible. Our approach is validated in simulation and real robot experiments on a 7DoF Franka Research 3 manipulator.

Constrained Passive Interaction Control: Leveraging Passivity and Safety for Robot Manipulators

TL;DR

This work addresses safe, passive physical interaction for torque-controlled manipulators in dynamic human environments by constraining a DS-based passive impedance controller with joint-space safety via exponential CBFs. It introduces ECBF formulations for hard joint-limit and self-collision constraints, and soft external-collision and singularity constraints using learnable neural boundary functions (SCA NN, Neural-JSDF) and a neural JSDF model, all integrated into a Relaxed-CBF-QP that prioritizes safety while preserving passivity whenever feasible. The approach is demonstrated in simulation and on a 7-DoF Franka manipulator, achieving constraint satisfaction and passive interaction under perturbations, with real-time performance on standard optimization tools. The framework holds promise for robust, safe manipulation in human-centric settings and highlights avenues to improve efficiency, such as more compact boundary-function representations and broader hardware deployments.

Abstract

Passivity is necessary for robots to fluidly collaborate and interact with humans physically. Nevertheless, due to the unconstrained nature of passivity-based impedance control laws, the robot is vulnerable to infeasible and unsafe configurations upon physical perturbations. In this paper, we propose a novel control architecture that allows a torque-controlled robot to guarantee safety constraints such as kinematic limits, self-collisions, external collisions and singularities and is passive only when feasible. This is achieved by constraining a dynamical system based impedance control law with a relaxed hierarchical control barrier function quadratic program subject to multiple concurrent, possibly contradicting, constraints. Joint space constraints are formulated from efficient data-driven self- and external C^2 collision boundary functions. We theoretically prove constraint satisfaction and show that the robot is passive when feasible. Our approach is validated in simulation and real robot experiments on a 7DoF Franka Research 3 manipulator.
Paper Structure (27 sections, 30 equations, 3 figures)

This paper contains 27 sections, 30 equations, 3 figures.

Table of Contents

  1. INTRODUCTION
  2. PROBLEM STATEMENT
  3. PRELIMINARIES
  4. Dynamical System Motion Planning
  5. Passivity-Based Control with Dynamical Systems kronander2015passive
  6. QP-based Controller via Control Barrier Functions ames2019cbf
  7. PROPOSED APPROACH
  8. Exponential Control Barrier Function Primer
  9. Joint Limit Avoidance Constraint
  10. Self-Collision Avoidance Constraint
  11. Model description
  12. Constraint formulation
  13. External Object Collision Avoidance Constraint
  14. Brief Introduction to JSDF
  15. Constraint formulation
  16. ...and 12 more sections

Figures (3)

  • Figure 1: Schematic of our proposed constrained passive interaction control framework. Green blocks denote novel components. Note that every block has a feedback term, from joint-space to task-space layers.
  • Figure 2: Evolution of joint-space boundary functions with active and inactive constraints on the 7-DoF Franka Research 3 manipulator for the simulated tests in Section \ref{['section5a']}. (left) Evolution $h_{SCA}$ and the minimal distance between link1 and link6 of a. (center) Minimal distance between the robot and the external ball with and without the external object collision constraints. Only link6 and link5 are reported, which tend to collide with the ball that we defined without the constraints. (right) Evolution of Manipulability Index $\mathcal{MI}(q)$.
  • Figure 3: (left) Evolution of all joint-space boundary functions for the simulated tests in Section \ref{['section5a']}. (center) Controller on real-robot allowing external perturbations yet satisfying self-collision and joint limit constraints. (right) $F_{ext}$ and $h_{SCA}$ for real-robot experiment.

Theorems & Definitions (2)

  • Definition 1: Set Invariance
  • Remark 1: Passivity