A 'MAP' to find high-performing soft robot designs: Traversing complex design spaces using MAP-elites and Topology Optimization

TL;DR

The paper tackles the lack of robust design tools for soft robotics by introducing Optimized Initial Design Domain (OIDD), a hybrid approach that merges MAP-Elites quality-diversity search with topology optimization. By embedding void regions and evolving their configuration, OIDD expands the TO design space, yielding diverse, high-performing structures while maintaining manufacturability through physics-based SIMP solvers. Demonstrations on a benchmark MBB beam show statistically significant improvements over standard SIMP, and applications to soft grippers and pneumatic fingers reveal rich design diversity and practical grasping performance, including experimental validation. The work provides a principled framework for generating innovative soft-robot designs in complex domains and highlights the importance of evolving the design domain itself to escape local optima and explore novel configurations.

Abstract

Soft robotics has emerged as the standard solution for grasping deformable objects, and has proven invaluable for mobile robotic exploration in extreme environments. However, despite this growth, there are no widely adopted computational design tools that produce quality, manufacturable designs. To advance beyond the diminishing returns of heuristic bio-inspiration, the field needs efficient tools to explore the complex, non-linear design spaces present in soft robotics, and find novel high-performing designs. In this work, we investigate a hierarchical design optimization methodology which combines the strengths of topology optimization and quality diversity optimization to generate diverse and high-performance soft robots by evolving the design domain. The method embeds variably sized void regions within the design domain and evolves their size and position, to facilitating a richer exploration of the design space and find a diverse set of high-performing soft robots. We demonstrate its efficacy on both benchmark topology optimization problems and soft robotic design problems, and show the method enhances grasp performance when applied to soft grippers. Our method provides a new framework to design parts in complex design domains, both soft and rigid.

Figures (9)

• Figure 1: Example of high-performing soft robotic grippers generated through OIDD method.
• Figure 2: Illustration of the encoded design domain initialization. a) 2D Design Domain Initialization b) 3D Design Domain Initialization
• Figure 3: Loading and boundary conditions of MBB beam. Left: full design domain, right: half design domain with symmetry boundary conditions.
• Figure 4: Analysis of MBB Beam Optimization. a) Evolution of archive coverage and corresponding optimal designs. b) Distribution of high-performing solutions in the behavior space of MBB beam
• Figure 5: Design domain of soft gripper optimization: the finger should bend inwards from the bottom left corner when a force is applied at the top right. The void space is intended to maintain a clear path for the gripper to close in a 2 fingered configuration