Machine Learning for Underwater Robotic Manipulation

为水下机器人操作开发有效的和鲁棒的控制，该项目的范围包括集成机器人系统的同时控制，机器人和机械手的多模态感觉伺服，如视觉和力/力矩伺服控制，环境视觉和声纳数据的集成，以及开发智能人机交互系统，用于在远程人类命令下的半自主机器人操作。该项目的重点将是应用机器学习技术来解决范围内的问题。

Develop effective and robust control for underwater robotic manipulation with an underwater vehicle and a robotic arm attached to it. The scope of the project includes simultaneous control of the integrated robotic system, multi-modality sensorial servo-ing of the vehicle and manipulator, such as with visual and force/torque servo control, integration of environmental visual and sonar data for manipulation, and developing intelligent human-robot interaction system for semi-autonomous robotic manipulation under remote human commands. The focus of the project will be the application of machine learning techniques to address the problems within the scope.