arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1703.04550cs.ROcs.LGcs.NEcs.SYeess.SY
收藏

通过深度强化学习实现机器人控制的传感器融合

Sensor Fusion for Robot Control through Deep Reinforcement Learning

Ghent University - imec, IDLab, Department of Information Technology(根特大学 - imec,IDLab,信息技术系)

Steven Bohez, Tim Verbelen, Elias De Coninck, Bert Vankeirsbilck, Pieter Simoens, Bart Dhoedt

AI总结:

本文提出通过深度强化学习实现机器人传感器信息融合,提升机器人在搜索和拾取任务中的鲁棒性和性能。

摘要:

深度强化学习正日益成为机器人控制算法的热门方法,旨在使机器人能够从非结构化感官输入中自学习有用的特征表示,从而获得最优的操作策略。除了机器人上的传感器外,环境中的传感器也可能被部署,尽管这些可能需要通过不可靠的无线连接访问。在本文中,我们展示了能够融合多个传感器信息并具有运行时传感器故障鲁棒性的深度神经网络架构。我们评估了我们的方法在机器人搜索和拾取任务中的性能,包括仿真和现实世界中的测试。

英文摘要:

Deep reinforcement learning is becoming increasingly popular for robot control algorithms, with the aim for a robot to self-learn useful feature representations from unstructured sensory input leading to the optimal actuation policy. In addition to sensors mounted on the robot, sensors might also be deployed in the environment, although these might need to be accessed via an unreliable wireless connection. In this paper, we demonstrate deep neural network architectures that are able to fuse information coming from multiple sensors and are robust to sensor failures at runtime. We evaluate our method on a search and pick task for a robot both in simulation and the real world.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑