Learning Purposeful Co-designs of Bio-inspired Quadrupedal Robots
Learning Purposeful Co-designs of Bio-inspired Quadrupedal Robots
This paper introduces Dogger, a bio-inspired quadruped featuring modular three-segment legs, four-bar linkages, and adjustable passive knee compliance. The authors propose a two-stage framework combining reinforcement learning with Bayesian optimization to train morphology-conditioned control policies across designs, which are then fixed during task-specific design selection. Deployed zero-shot, the resulting designs reduce cost of transport by 18.6% for inspection and 35.7% for payload-carrying tasks compared to baselines. The work demonstrates that expanding physical design space alongside learned control is essential for realizing task-driven morphology-control co-design in legged robotics.
The study presents Dogger, a bio-inspired quadrupedal robot with modular legs and adjustable compliance, evaluated via a reinforcement learning and Bayesian optimization framework that achieves reported cost-of-transport reductions of 18.6% and 35.7% for inspection and payload tasks respectively.
Source: arXiv Robotics — research abstracts · Read original article ↗
Loading article text…
What the source reports
Publisher-reported claims, with original evidence. These results have not been independently verified by RoboSignal.
- paper: not_reported
Source excerpts and review record
Automatically extracted; no manual editorial approval recorded.
iance. 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-com
Open source S3
arXiv:2611.02462v1 Announce Type: new 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 compl
Open source S2
pliance 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.
Open source S4
Source：arXiv Robotics — research abstracts · arxiv.org