Bio-inspired robot legs designed with smarter control improve efficiency
Learning Purposeful Co-designs of Bio-inspired Quadrupedal Robots
Robotics
Summary
Designing robot legs and how they move separately can limit how well robots work. The authors created a robot dog with special flexible legs that can be adjusted in many ways. They combined smart learning methods to find the best design and control for different tasks. The improved designs made the robot use less energy when inspecting areas or carrying loads. This shows that designing a robot’s body and brain together leads to better results.
What this means in practice
- •For robotics engineers: Design and build quadruped robots with modular legs and adaptive control to optimize energy efficiency for inspection and payload tasks.
- •For field robotics operators: Deploy legged robots with task-specific designs that reduce energy use in inspection or carrying payloads without extra retraining.
Authors
Fabio Boekel, Jiatao Ding, Cosimo Della Santina
Abstract
Treating morphology and control as separate design problems has constrained the efficiency and adaptability of quadrupedal robots, while current co-design approaches are often constrained by restricted design spaces. We overcome this limit by introducing Dogger, a bio-inspired quadruped with modular three-segment legs, four-bar linkages and adjustable passive knee compliance. A two-stage framework combining reinforcement learning and Bayesian optimization is developed: the morphology-conditioned control policy is trained across designs and then held fixed during task-specific design selection. The resulting designs and policies deployed zero-shot reduce the cost of transport by 18.6% and 35.7% for inspection and payload-carrying tasks. Comparisons demonstrate that full geometry-compliance co-design outperforms fixed and restricted design scopes. Cross-task tests further reveal distinct task-specific morphologies with shared mechanical trade-offs. These results provide hardware-level evidence that expanding the physical design space alongside learned control is essential to realize the full potential of task-driven morphology-control co-design in legged robotics.