Learning of Parameters in Behavior Trees for Movement Skills

TL;DR

This paper introduces a novel algorithm that learns Behavior Tree parameters in simulation for robotic movement skills, enabling transfer to real robots without further training, thus improving safety and interpretability.

Contribution

The paper presents a new method for optimizing Behavior Tree parameters in simulation that generalizes to real robots without additional training, enhancing safety and interpretability.

Findings

Outperforms baseline methods in obstacle avoidance tasks

Successfully transfers learned parameters from simulation to real robot

Demonstrates effectiveness in contact-rich insertion tasks

Abstract

Reinforcement Learning (RL) is a powerful mathematical framework that allows robots to learn complex skills by trial-and-error. Despite numerous successes in many applications, RL algorithms still require thousands of trials to converge to high-performing policies, can produce dangerous behaviors while learning, and the optimized policies (usually modeled as neural networks) give almost zero explanation when they fail to perform the task. For these reasons, the adoption of RL in industrial settings is not common. Behavior Trees (BTs), on the other hand, can provide a policy representation that a) supports modular and composable skills, b) allows for easy interpretation of the robot actions, and c) provides an advantageous low-dimensional parameter space. In this paper, we present a novel algorithm that can learn the parameters of a BT policy in simulation and then generalize to the physical robot without any additional training. We leverage a physical simulator with a digital twin of our workstation, and optimize the relevant parameters with a black-box optimizer. We showcase the efficacy of our method with a 7-DOF KUKA-iiwa manipulator in a task that includes obstacle avoidance and a contact-rich insertion (peg-in-hole), in which our method outperforms the baselines.