The major challenges of collision avoidance for robot navigation in crowded scenes lie in accurate environment modeling, fast perceptions, and trustworthy motion planning policies. This paper presents a novel adaptive environment model based collision avoidance reinforcement learning (i.e., AEMCARL) framework for an unmanned robot to achieve collision-free motions in challenging navigation scenarios. The novelty of this work is threefold: (1) developing a hierarchical network of gated-recurrent-unit (GRU) for environment modeling; (2) developing an adaptive perception mechanism with an attention module; (3) developing an adaptive reward function for the reinforcement learning (RL) framework to jointly train the environment model, perception function and motion planning policy. The proposed method is tested with the Gym-Gazebo simulator and a group of robots (Husky and Turtlebot) under various crowded scenes. Both simulation and experimental results have demonstrated the superior performance of the proposed method over baseline methods.

Department of Labor)-backed mine safety mission – achieved a historic unmanned underground mine inspection at one of the US' largest underground room and pillar limestone operations in this comprehensive IM report. Using ten ADR Explora XL unmanned robots, a Rajant wireless Kinetic Mesh below-ground communication network, and PBE hardware and technology, a horizontal mobile infrastructure distance of 1.7 km was achieved. This allowed the unmanned robots to record high-definition video and LiDAR to create a virtual 3D mine model to assess the condition of the mine, for the deepest remote underground mine inspection in history. The inspection made it possible for MSHA to conclude within a very short time that it was safe to re-enter the operation and begin remediation efforts, which included allowing mine personnel back into the mine to re-establish power and communications, after which mining was able to recommence quickly at the site. The project, in many ways, is the ultimate example of necessity breeding innovation.

Idaho has become the second U.S. state to pass legislation to permit unmanned, ground-based delivery robots to rove around on sidewalks across the state. Earlier this month, Virginia made robotics history as the first state to pass a law specifically addressing the use of autonomous terrestrial delivery robots. The new Idaho law, which was signed by the governor today, goes into effect July 1. The legislation was championed by state Republican lawmakers, Jason Monks in the House and Bert Brackett in the Senate. Monks worked with Starship Technologies, an Estonia-based robot delivery company, on the legislation, which sailed through both state houses to pass in less than a month.