We develop an algorithm to control an underactuated unmanned surface vehicle (USV) using kinodynamic motion planning with funnel control (KDF). KDF has two key components: motion planning used to generate trajectories with respect to kinodynamic constraints, and funnel control, also referred to as prescribed performance control, which enables trajectory tracking in the presence of uncertain dynamics and disturbances. We extend prescribed performance control to address the challenges posed by underactuation and control-input saturation present on the USV. The proposed scheme guarantees stability under user-defined prescribed performance functions where model parameters and exogenous disturbances are unknown. Furthermore, we present an optimization problem to obtain smooth, collision-free trajectories while respecting kinodynamic constraints. We deploy the algorithm on a USV and verify its efficiency in real-world open-water experiments.

We propose a control protocol based on the prescribed performance control (PPC) methodology for a quadro-tor unmanned aerial vehicle (UAV). Quadrotor systems belong to the class of underactuated systems for which the original PPC methodology cannot be directly applied. We introduce the necessary design modifications to stabilize the considered system with prescribed performance. The proposed control protocol does not use any information of dynamic model parameters or exogenous disturbances. Furthermore, the stability analysis guarantees that the tracking errors remain inside of designer-specified time-varying functions, achieving prescribed performance independent from the control gains’ selection. Finally, simulation results verify the theoretical results.

The paper considers a problem of 3D environment model reconstruction from a set of 2D images acquired by the Unmanned Aerial Vehicle (UAV) in near real-time. The designed framework combines the FAST (Features from Accelerated Segment Test) algorithm and optical flow approach for detection of interest image points and adjacent images reconstruction. The robust estimation of camera locations is performed using the image points tracking. The coordinates of 3D points and the projection matrix are computed simultaneously using Structure-from-Motion (SfM) algorithm, from which the 3D model of environment is generated. The designed framework is tested using real image data and video sequences captured with camera mounted on the UAV. The effectiveness and quality of the proposed framework are verified through analyses of accuracy of the 3D model reconstruction and its time execution.

Deep learning and reinforcement learning have had dramatic recent breakthroughs. However, the ability to apply these approaches to control real physically embodied agents remains primitive compared to traditional robotics approaches. To help bridge this gap, we are announcing the AI Driving Olympics (AI-DO), which will be a live competition at the Neural Information Processing Systems (NIPS) in Dec. 2018. The overall objective of the competition is to evaluate the state of the art of machine learning and artificial intelligence on a physically embodied platform. We are using the Duckietown [14] platform since it is a simple and well-specified environment that can be used for autonomous navigation. The competition comprises five tasks of increasing complexity - from simple lane following to managing an autonomous fleet. For each task we will provide tools for competitors to use in the form of simulators, logs, low-cost access to robotic hardware and