Innate Motivation for Robot Swarms by Minimizing Surprise: From Simple Simulations to Real-World Experiments

TL;DR

This paper explores innate motivation in robot swarms by minimizing surprise through sensor prediction accuracy, demonstrating successful transfer from simulations to real-world robots and highlighting benefits for robustness and scalability.

Contribution

It introduces a surprise-minimization approach for innate motivation in swarm robots, validated through simulations and real-world experiments, addressing challenges of reward design.

Findings

Enhanced behavioral diversity and robustness in swarm behaviors

Scalability of evolved behaviors confirmed in larger robot groups

Successful transfer of simulation results to real robots

Abstract

Applications of large-scale mobile multi-robot systems can be beneficial over monolithic robots because of higher potential for robustness and scalability. Developing controllers for multi-robot systems is challenging because the multitude of interactions is hard to anticipate and difficult to model. Automatic design using machine learning or evolutionary robotics seem to be options to avoid that challenge, but bring the challenge of designing reward or fitness functions. Generic reward and fitness functions seem unlikely to exist and task-specific rewards often have undesired side effects. Approaches of so-called innate motivation try to avoid the specific formulation of rewards and work instead with different drivers, such as curiosity. Our approach to innate motivation is to minimize surprise, which we implement by maximizing the accuracy of the swarm robot's sensor predictions using neuroevolution. A unique advantage of the swarm robot case is that swarm members populate the robot's environment and can trigger more active behaviors in a self-referential loop. We summarize our previous simulation-based results concerning behavioral diversity, robustness, scalability, and engineered self-organization, and put them into context. In several new studies, we analyze the influence of the optimizer's hyperparameters, the scalability of evolved behaviors, and the impact of realistic robot simulations. Finally, we present results using real robots that show how the reality gap can be bridged.