Skip to main content

Compliant Control & Force-Limiting Architecture

Duration: 55 min · Level: Intermediate · Module: 8. Safety & Human-Robot Interaction · Focus: compliance, force-control, impedance, safety

Learning objectives

By the end of this lesson you will be able to explain and apply:

  • Joint torque sensing
  • Wrist F/T sensors
  • Impedance control
  • Collision detection without external sensors
  • Predictive safety

Why this matters​

Meeting ISO/TS 15066 force limits requires that G1 can sense and limit contact forces in real-time.

Overview​

Meeting ISO/TS 15066 force limits requires that G1 can sense and limit contact forces in real-time. This lesson covers the hardware and software architecture for force-limited operation — from actuator torque sensing to predictive collision detection.

Key concepts​

Key idea

Joint torque sensing: strain gauges or current sensing at each motor enable contact force estimation; accuracy ±2-5N with proprioceptive sensing only

  • Wrist F/T sensors: 6-DOF force/torque measurement at each wrist; provides accurate end-effector force for manipulation safety; ATI Mini45 or custom designs
  • Impedance control: regulate robot endpoint stiffness/damping rather than position; "soft" impedance (low stiffness) makes contact gentle; used in Franka Emika Panda cobots
  • Collision detection without external sensors: use momentum observer (estimated vs actual joint velocities) to detect unexpected contact; detects collisions in <10ms
  • Predictive safety: trajectory planning checks all future configurations against workspace model; stops motion before entering known occupied region
  • ISO 10218-2 risk assessment: document all hazard scenarios, estimate probability and severity, implement mitigations until residual risk is acceptable

Check your understanding​

Try to recall each answer before expanding it.

Q1. What do you know about Joint torque sensing?

strain gauges or current sensing at each motor enable contact force estimation; accuracy ±2-5N with proprioceptive sensing only

Q2. What do you know about Wrist F/T sensors?

6-DOF force/torque measurement at each wrist; provides accurate end-effector force for manipulation safety; ATI Mini45 or custom designs

Q3. What do you know about Impedance control?

regulate robot endpoint stiffness/damping rather than position; "soft" impedance (low stiffness) makes contact gentle; used in Franka Emika Panda cobots

Q4. What do you know about Collision detection without external sensors?

use momentum observer (estimated vs actual joint velocities) to detect unexpected contact; detects collisions in <10ms

Q5. What do you know about Predictive safety?

trajectory planning checks all future configurations against workspace model; stops motion before entering known occupied region

← Previous: 8.1 ISO 10218 & ISO/TS 15066: The Collaborative Robot Safety Standards · Next: 8.3 Behavior Trees & Task-Level Safety →

Part of Module 8: Safety & Human-Robot Interaction.