用价值梯度优化机器人形态，一次训练多形态适配。
Shape Your Body: Value Gradients for Multi-Embodiment Robot Design
- 训练通用价值函数后，通过梯度反向优化新机器人形态。
- 在50个机器人、超1100维参数空间上实现跨形态优化。
- 可定位性能瓶颈，适合机器人设计与分析场景。
我们提出将通用多形态价值函数转化为可复用的机器人设计模型。无需为每个机器人重新运行强化学习联合设计流程，我们先在多种机器人形态上训练具备形态感知能力的策略与价值函数。训练完成后，冻结的价值函数作为可微代理，通过价值梯度优化候选形态。我们在不同设计场景下评估该方法，涵盖单机器人扰动及跨形态类别的未见机器人，模型在最多50个机器人、超过1100个连续形态参数的设计空间上训练。除完整形态优化外，还证明价值梯度可识别影响性能的关键设计与控制参数，支持新机器人设计的优化与分析。
原文摘要 · Abstract (English)
We propose to turn generalist multi-embodiment value functions into reusable models for robot design. Instead of running a new reinforcement learning co-design loop for each robot, we first train an embodiment-aware policy and value function across many robot designs. After training, the frozen value function is used as a differentiable surrogate to optimize candidate embodiments through value gradients. We evaluate our approach across different robot design settings, from perturbed single robots to held-out robots across morphology classes, with single models trained on up to 50 robots and design spaces of over 1100 continuous embodiment parameters. Beyond optimizing complete embodiments, we show that value gradients can identify performance-limiting design and control parameters, enabling both the optimization and the analysis of new robot designs.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。