Repairing Neural Networks for Safety in Robotic Systems using Predictive Models

TL;DR

This paper presents a safety-aware robot learning method that repairs policies using predictive models, combining behavioral cloning with neural network repair to ensure safety constraints are met in navigation and prosthetic applications.

Contribution

It introduces a two-step supervised learning framework that integrates predictive models for safety-aware policy repair in robotic systems.

Findings

Policies adhere to safety constraints in navigation and prosthetic tasks

Method reduces interaction time with robots significantly

Effective in real-world robotic safety applications

Abstract

This paper introduces a new method for safety-aware robot learning, focusing on repairing policies using predictive models. Our method combines behavioral cloning with neural network repair in a two-step supervised learning framework. It first learns a policy from expert demonstrations and then applies repair subject to predictive models to enforce safety constraints. The predictive models can encompass various aspects relevant to robot learning applications, such as proprioceptive states and collision likelihood. Our experimental results demonstrate that the learned policy successfully adheres to a predefined set of safety constraints on two applications: mobile robot navigation, and real-world lower-leg prostheses. Additionally, we have shown that our method effectively reduces repeated interaction with the robot, leading to substantial time savings during the learning process.