Object Rearrangement Using Learned Implicit Collision Functions
Object Rearrangement Using Learned Implicit Collision Functions
复制标题
DOI：
10.1109/icra48506.2021.9561516
复制
发表时间：
2020-11
期刊：
影响因子：
--
通讯作者：
Michael Danielczuk;Arsalan Mousavian;Clemens Eppner;D. Fox
中科院分区：
文献类型：
--
作者：
Michael Danielczuk;Arsalan Mousavian;Clemens Eppner;D. Fox
机器人物体重新排列结合了拾取和放置物体的技能。当物体模型不可用时，典型的碰撞检测模型可能无法预测存在遮挡的局部点云中的碰撞，这使得生成无碰撞抓取或放置轨迹具有挑战性。我们提出一种学习到的碰撞模型，该模型接受场景和查询物体的点云，并预测场景内6自由度物体姿态的碰撞情况。我们使用100万个场景/物体点云对和20亿个碰撞查询的合成数据集来训练该模型。我们在桌面重新排列任务中利用学习到的碰撞模型作为模型预测路径积分（MPPI）策略的一部分，并表明该策略能够为在模拟和实际杂乱场景中训练时未见过的物体使用Franka Panda机器人规划无碰撞的抓取和放置。在模拟碰撞查询数据集上，学习到的模型在准确性方面比传统流程和学习到的简化模型高出9.8%，并且比性能最佳的基线快75倍。视频和补充材料可在https://research.nvidia.com/publication/2021 - 03_Object - Rearrangement - Using获取。
Robotic object rearrangement combines the skills of picking and placing objects. When object models are unavailable, typical collision-checking models may be unable to predict collisions in partial point clouds with occlusions, making generation of collision-free grasping or placement trajectories challenging. We propose a learned collision model that accepts scene and query object point clouds and predicts collisions for 6DOF object poses within the scene. We train the model on a synthetic set of 1 million scene/object point cloud pairs and 2 billion collision queries. We leverage the learned collision model as part of a model predictive path integral (MPPI) policy in a tabletop rearrangement task and show that the policy can plan collision-free grasps and placements for objects unseen in training in both simulated and physical cluttered scenes with a Franka Panda robot. The learned model outperforms both traditional pipelines and learned ablations by 9.8% in accuracy on a dataset of simulated collision queries and is 75x faster than the best-performing baseline. Videos and supplementary material are available at https://research.nvidia.com/publication/2021-03_Object-Rearrangement-Using.