Online Aerial Terrain Mapping for Ground Robot Navigation.

This work presents a collaborative unmanned aerial and ground vehicle system which utilizes the aerial vehicle's overhead view to inform the ground vehicle's path planning in real time. The aerial vehicle acquires imagery which is assembled into a orthomosaic and then classified. These terrain classes are used to estimate relative navigation costs for the ground vehicle so energy-efficient paths may be generated and then executed. The two vehicles are registered in a common coordinate frame using a real-time kinematic global positioning system (RTK GPS) and all image processing is performed onboard the unmanned aerial vehicle, which minimizes the data exchanged between the vehicles. This paper describes the architecture of the system and quantifies the registration errors between the vehicles.

Improving Autonomous Robotic Navigation Using Imitation Learning.

Autonomous navigation to a specified waypoint is traditionally accomplished with a layered stack of global path planning and local motion planning modules that generate feasible and obstacle-free trajectories. While these modules can be modified to meet task-specific constraints and user preferences, current modification procedures require substantial effort on the part of an expert roboticist with a great deal of technical training. In this paper, we simplify this process by inserting a Machine Learning module between the global path planning and local motion planning modules of an off-the shelf navigation stack. This model can be trained with human demonstrations of the preferred navigation behavior, using a training procedure based on Behavioral Cloning, allowing for an intuitive modification of the navigation policy by non-technical users to suit task-specific constraints. We find that our approach can successfully adapt a robot's navigation behavior to become more like that of a demonstrator. Moreover, for a fixed amount of demonstration data, we find that the proposed technique compares favorably to recent baselines with respect to both navigation success rate and trajectory similarity to the demonstrator.