Bio-inspired stress-sharing enables autonomous modular spacecraft repair

A new paper from researchers Sidhdharth D. Sikka, Yue Shen, and Shaoshuai Mou presents a bio-inspired 'stress-sharing' approach to fault repair in modular spacecraft. The problem: structural damage in such spacecraft can break mechanical and communication links, reducing system capability. Existing solutions rely on redundancy or preplanned reconfiguration, which don't scale or adapt autonomously. The team models the spacecraft as a lattice-constrained graph and introduces a decentralized rule set: when damage occurs, modules emit local distress signals, and surviving modules move toward the damaged area to close gaps, mimicking wound healing. After consolidation, each module locally retraces its own movements to restore the original shape—all without absolute position sensing or central commands.

To test effectiveness, the team ran rigid-body simulations in PyBullet on structures ranging from 10 to 160 modules, with fault densities of 10%, 20%, and 30% for both random and localized damage. Results were strong: even in the most severe case—30% random module failures—the policy gathered roughly 80% of surviving modules into a single connected component. Importantly, this fraction improved with assembly size, making the approach scalable for large swarms. The method's reliance on only local information and asynchronous operation suits deep-space environments where communication lags and global awareness are impractical. The work is submitted to IEEE Transactions on Aerospace and Electronic Systems and suggests a viable path toward resilient, self-repairing spacecraft for long-duration missions.