SABER：基于注意力的语义可供性学习用于腿式运动
SABER: Learning Attention-based Semantic Affordance for Legged Locomotion
- Institute of Advanced Intelligence and Computing (IAIC), Agency for Science, Technology and Research (A*STAR)（先进智能与计算研究所，新加坡科技研究局）
- Nanyang Technological University（南洋理工大学）
机构由 AI 辅助整理，请以论文原文为准。
中文总结 AI 辅助
SABER提出一种无需规划器的强化学习策略，通过结合地形几何与语义接触成本，利用带语义偏置的交叉注意力选择安全落脚点，在宇树B2上实现仿真到现实的语义接触选择，显著减少禁止接触。
中文摘要 AI 辅助
感知腿式运动通过将地形几何信息整合到学习策略中已取得快速发展，然而地形语义的整合仍然稀疏：管道、草地或易碎箱子可能在几何上可通行，但接触它们却不合适。在腿式机器人日益广泛应用的工业环境中，一步踏错可能损坏易碎设备、使机器人失稳或危及现场安全。为解决这一问题，我们提出了SABER，一种无需规划器的强化学习策略，它联合推理地形几何和语义接触许可。该策略使用统一的地形可供性地图，其中每个单元编码局部3D几何和语义接触成本。我们通过一种学习到的有符号语义偏置增强交叉注意力：该偏置是注意力logits上的加性项，由接触成本门控，根据标记单元与最近脚的距离重新加权。因此，危险区域在仍能影响下一步落脚点的地方重塑注意力，而在无法影响的地方其影响逐渐消失。由此产生的策略在允许的支撑面上选择落脚点，并在摆动阶段保持腿部远离禁止区域。我们进行了系统的消融实验，隔离每个架构组件的贡献；仅移除语义偏置就会使禁止接触增加55%，而速度跟踪保持不变。我们在宇树B2上验证了该策略，展示了在室内和室外环境以及四种语义障碍类别上的仿真到现实语义接触选择。
英文摘要
Perceptive legged locomotion has advanced rapidly by integrating terrain geometry into learned policies, yet the integration of terrain meaning remains sparse: a pipe, a patch of grass, or a fragile box may be geometrically traversable while being inappropriate for contact. In industrial environments, where legged robots increasingly operate, a single misplaced step can damage fragile equipment, destabilize the robot, or endanger the site. To address this, we introduce SABER, a planner-free reinforcement-learning policy that jointly reasons about terrain geometry and semantic contact permission. The policy consumes a unified terrain-affordance map, where each cell encodes local 3D geometry and a semantic contact cost. We augment cross-attention with a learned, signed semantic bias: an additive term on the attention logits, gated by the contact cost, that reweights flagged cells by their distance from the nearest foot. A hazard therefore reshapes attention where it can still affect the next foothold, and its influence fades where it cannot. The resulting policy selects footholds on permitted support and keeps the leg clear of forbidden regions throughout the swing phase. We perform a systematic ablation that isolates the contribution of each architectural component; removing the semantic bias alone increases forbidden contacts by 55% while velocity tracking is unchanged. We validate the policy on a Unitree B2, demonstrating sim-to-real semantic contact selection across indoor and outdoor environments and four semantic obstacle classes.
