Robot co-design boosts walking skills with actor inheritance and collaboration
LACE-CRAFT: Robot Co-Design with Actor Inheritance and Blackboard Collaboration
Robotics
Summary
Designing robots usually means starting over each time, which wastes learning about how to move. The authors introduce LACE-CRAFT, a method that reuses what a robot already learned when trying new body shapes and goals, speeding up training. Their system also uses a collaboration method to create and review robot designs together, improving quality. Experiments show their approach outperforms previous methods on walking tasks, and they even built a real walking robot prototype to demonstrate it.
What this means in practice
- •For robotics engineers: Generate and improve robot body designs and control policies simultaneously to create more effective walking robots faster.
- •For robotics hardware developers: Use a geometry-to-hardware workflow to convert simulated robot designs into physical prototypes with configured joints and actuators.
Authors
Yuhan Wen, Jiawei Wang, Qixuan Zhang, Yusen Qin, Lan Xu
Abstract
Robot co-design couples morphology search with policy learning, yet training every new design from scratch discards acquired control experience. We present LACE-CRAFT, which compares continued learning on the current robot with policy adaptation to new morphology-reward pairs. LACE resumes the incumbent's full learning state and initializes compatible challengers with its actor parameters and observation statistics. A fixed task metric selects among both branches and the frozen incumbent. CRAFT coordinates Feedback, Morphology, Reward, and Integration roles through shared experimental records and behavioral replays to generate and cross-review paired proposals. A generative extension converts generated meshes into editable articulated models with configured joints, actuator interfaces, and consistently updated simulation assets. Across five locomotion benchmarks, mean scores over three evaluation seeds are 6.4-91.9% higher than D2C. Both methods train 30 new morphology-reward pairs over five rounds; LACE additionally uses four continuation training units. Five-task ablations examine policy inheritance and replay-derived feedback. A fabricated prototype demonstrates indoor walking and illustrates the geometry-to-hardware workflow.