NeuroEvolution algorithms applied in the designing process of biohybrid actuators

TL;DR

The study tackles automated design of biohybrid soft robots by evaluating NEAT and HyperNEAT against AFPO for 3D biohybrid morphologies encoded via neural networks. Using a client-server workflow and Voxelyze physics simulations, it compares performance on maximum displacement and the trade-off between morphology volume and displacement. HyperNEAT tends to produce compact, robust morphologies with displacement comparable to or exceeding NEAT, while NEAT often yields higher maxima and tighter robustness, all outperforming AFPO in several scenarios. The findings suggest HyperNEAT as a particularly promising approach for efficient, manufacturable BHMs with potential biomedical applications, enabled by distributed computation to manage heavy simulation workloads.

Abstract

Soft robots diverge from traditional rigid robotics, offering unique advantages in adaptability, safety, and human-robot interaction. In some cases, soft robots can be powered by biohybrid actuators and the design process of these systems is far from straightforward. We analyse here two algorithms that may assist the design of these systems, namely, NEAT (NeuroEvolution of Augmented Topologies) and HyperNEAT (Hypercube-based NeuroEvolution of Augmented Topologies). These algorithms exploit the evolution of the structure of actuators encoded through neural networks. To evaluate these algorithms, we compare them with a similar approach using the Age Fitness Pareto Optimization (AFPO) algorithm, with a focus on assessing the maximum displacement achieved by the discovered biohybrid morphologies. Additionally, we investigate the effects of optimization against both the volume of these morphologies and the distance they can cover. To further accelerate the computational process, the proposed methodology is implemented in a client-server setting; so, the most demanding calculations can be executed on specialized and efficient hardware. The results indicate that the HyperNEAT-based approach excels in identifying morphologies with minimal volumes that still achieve satisfactory displacement targets.

Figures (8)

• Figure 1: Substrate utilised under HyperNEAT for morphology generator.
• Figure 2: Diagram of the implementation of NEAT and HyperNEAT for producing BHM morphologies.
• Figure 3: Evaluation of individuals with processes triggered and concurrent interactions with Voxelyze instances.
• Figure 4: Displacement observed under 500 different phase offset scenarios for the controllers of the fittest morphologies found in five different evolutionary runs under: (a) AFPO-based approach, (b) NEAT-based approach, and (c) HyperNEAT-based approach.
• Figure 5: Fittest morphology in terms of reaching maximum displacement under: AFPO, NEAT, and HyperNEAT.