Learning Behavior Trees with Genetic Programming in Unpredictable Environments

TL;DR

This paper demonstrates that genetic programming can effectively learn behavior trees for robots operating in unpredictable environments, enabling adaptable, fault-tolerant solutions without task-specific heuristics.

Contribution

It introduces a method using genetic programming to learn behavior trees in simulation that transfer effectively to realistic environments, reducing manual tuning.

Findings

Learned behavior trees solve tasks in realistic simulators

Method is fault-tolerant and adaptable

No task-specific heuristics needed

Abstract

Modern industrial applications require robots to be able to operate in unpredictable environments, and programs to be created with a minimal effort, as there may be frequent changes to the task. In this paper, we show that genetic programming can be effectively used to learn the structure of a behavior tree (BT) to solve a robotic task in an unpredictable environment. Moreover, we propose to use a simple simulator for the learning and demonstrate that the learned BTs can solve the same task in a realistic simulator, reaching convergence without the need for task specific heuristics. The learned solution is tolerant to faults, making our method appealing for real robotic applications.