论文标题
单眼摄像机和单光束基于域的水下无碰撞导航与域随机化
Monocular Camera and Single-Beam Sonar-Based Underwater Collision-Free Navigation with Domain Randomization
论文作者
论文摘要
水下导航提出了一些挑战，包括非结构化的未知环境，缺乏可靠的定位系统（例如GPS）和可见度不佳。此外，水下机器人的优质障碍物检测传感器很少且昂贵。 RGB-D摄像机和LIDAR之类的许多传感器仅在空中工作。为了启用可靠的无地图水下导航，尽管存在这些挑战，我们还是根据单眼相机和固定的单光束回声响起器提出了一个低成本的端到端导航系统，该系统有效地将水下机器人导航到航路点，同时避免在附近的障碍物。我们提出的方法基于近端策略优化（PPO），该方法将当前相对目标信息，估计的深度图像，回声声读取和先前执行的操作以及以归一化规模输出3D机器人操作。在模拟中进行了端到端培训，在该模拟中，我们采用了域随机化（不同的水下条件和可见性），以学习针对噪音和可见性条件变化的强大政策。仿真和现实世界中的实验表明，我们提出的方法在未知的水下环境中导航低成本的水下机器人方面取得了成功和韧性。该实现可在https://github.com/dartmouthrobotics/deeprl-uw-robot-nevigation上公开获得。
Underwater navigation presents several challenges, including unstructured unknown environments, lack of reliable localization systems (e.g., GPS), and poor visibility. Furthermore, good-quality obstacle detection sensors for underwater robots are scant and costly; and many sensors like RGB-D cameras and LiDAR only work in-air. To enable reliable mapless underwater navigation despite these challenges, we propose a low-cost end-to-end navigation system, based on a monocular camera and a fixed single-beam echo-sounder, that efficiently navigates an underwater robot to waypoints while avoiding nearby obstacles. Our proposed method is based on Proximal Policy Optimization (PPO), which takes as input current relative goal information, estimated depth images, echo-sounder readings, and previous executed actions, and outputs 3D robot actions in a normalized scale. End-to-end training was done in simulation, where we adopted domain randomization (varying underwater conditions and visibility) to learn a robust policy against noise and changes in visibility conditions. The experiments in simulation and real-world demonstrated that our proposed method is successful and resilient in navigating a low-cost underwater robot in unknown underwater environments. The implementation is made publicly available at https://github.com/dartmouthrobotics/deeprl-uw-robot-navigation.
