Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Control of Biohybrid Actuators using NeuroEvolution

Hugo Alcaraz-Herrera, Michail-Antisthenis Tsompanas, Andrew Adamatzky, Igor Balaz

TL;DR

This work tackles the challenge of designing controllers for soft biohybrid actuators (BHAs) used in medical contexts by applying neuroevolution. It compares Neuroevolution of Augmenting Topologies (NEAT) and Hypercube-based NEAT (HyperNEAT) against a standard genetic algorithm (SGA) to induce upward bending in BHAs within a Voxelyze-based simulation, evaluating general performance, robustness across morphologies, and controller complexity. Results show NEAT and HyperNEAT significantly outperform SGA, with NEAT achieving the highest mean displacement on several morphologies; HyperNEAT also outperforms SGA but is generally slightly behind NEAT, likely due to substrate design limitations. The findings suggest NE-based approaches are effective for morphology- and material-aware control of BHAs, with potential implications for implantable medical devices like catheter-based drug-delivery systems; NEAT, in particular, offers a favorable balance of performance and controller simplicity for real-world deployment.

Abstract

In medical-related tasks, soft robots can perform better than conventional robots because of their compliant building materials and the movements they are able perform. However, designing soft robot controllers is not an easy task, due to the non-linear properties of their materials. Since human expertise to design such controllers is yet not sufficiently effective, a formal design process is needed. The present research proposes neuroevolution-based algorithms as the core mechanism to automatically generate controllers for biohybrid actuators that can be used on future medical devices, such as a catheter that will deliver drugs. The controllers generated by methodologies based on Neuroevolution of Augmenting Topologies (NEAT) and Hypercube-based NEAT (HyperNEAT) are compared against the ones generated by a standard genetic algorithm (SGA). In specific, the metrics considered are the maximum displacement in upward bending movement and the robustness to control different biohybrid actuator morphologies without redesigning the control strategy. Results indicate that the neuroevolution-based algorithms produce better suited controllers than the SGA. In particular, NEAT designed the best controllers, achieving up to 25% higher displacement when compared with SGA-produced specialised controllers trained over a single morphology and 23% when compared with general purpose controllers trained over a set of morphologies.

Control of Biohybrid Actuators using NeuroEvolution

TL;DR

This work tackles the challenge of designing controllers for soft biohybrid actuators (BHAs) used in medical contexts by applying neuroevolution. It compares Neuroevolution of Augmenting Topologies (NEAT) and Hypercube-based NEAT (HyperNEAT) against a standard genetic algorithm (SGA) to induce upward bending in BHAs within a Voxelyze-based simulation, evaluating general performance, robustness across morphologies, and controller complexity. Results show NEAT and HyperNEAT significantly outperform SGA, with NEAT achieving the highest mean displacement on several morphologies; HyperNEAT also outperforms SGA but is generally slightly behind NEAT, likely due to substrate design limitations. The findings suggest NE-based approaches are effective for morphology- and material-aware control of BHAs, with potential implications for implantable medical devices like catheter-based drug-delivery systems; NEAT, in particular, offers a favorable balance of performance and controller simplicity for real-world deployment.

Abstract

In medical-related tasks, soft robots can perform better than conventional robots because of their compliant building materials and the movements they are able perform. However, designing soft robot controllers is not an easy task, due to the non-linear properties of their materials. Since human expertise to design such controllers is yet not sufficiently effective, a formal design process is needed. The present research proposes neuroevolution-based algorithms as the core mechanism to automatically generate controllers for biohybrid actuators that can be used on future medical devices, such as a catheter that will deliver drugs. The controllers generated by methodologies based on Neuroevolution of Augmenting Topologies (NEAT) and Hypercube-based NEAT (HyperNEAT) are compared against the ones generated by a standard genetic algorithm (SGA). In specific, the metrics considered are the maximum displacement in upward bending movement and the robustness to control different biohybrid actuator morphologies without redesigning the control strategy. Results indicate that the neuroevolution-based algorithms produce better suited controllers than the SGA. In particular, NEAT designed the best controllers, achieving up to 25% higher displacement when compared with SGA-produced specialised controllers trained over a single morphology and 23% when compared with general purpose controllers trained over a set of morphologies.

Paper Structure

This paper contains 15 sections, 3 equations, 4 figures, 3 tables.

Table of Contents

  1. Introduction
  2. Background
  3. Neuroevolution
  4. NEAT
  5. HyperNEAT
  6. Experimental setup
  7. Voxelyze
  8. Configuration Scheme
  9. NEAT configuration
  10. HyperNEAT configuration
  11. NEAT vs HyperNEAT
  12. General performance
  13. Robustness
  14. Controller complexity
  15. Conclusions

Figures (4)

  • Figure 1: Example of an BHA simulated by Voxelyze.
  • Figure 2: Substrate utilised under HyperNEAT to design BHA controllers.
  • Figure 3: Mean general performance with $\pm 95$ confidence interval (shaded region) under SGA, NEAT, and HyperNEAT using: (a) BHA 1; (b) BHA 2; and (c) BHA 3.
  • Figure 4: Displacement observed in the $yz$ plane of the top nine BHAs induced by: top - SGA (left), NEAT (centre), and HyperNEAT (right); bottom (close up) - NEAT (left), and HyperNEAT (right).