Michael Watterson

Bio: Michael Watterson is an academic researcher from University of Pennsylvania. The author has contributed to research in topics: Kalman filter & Trajectory optimization. The author has an hindex of 13, co-authored 19 publications receiving 849 citations.

Planning Dynamically Feasible Trajectories for Quadrotors Using Safe Flight Corridors in 3-D Complex Environments

Abstract: There is extensive literature on using convex optimization to derive piece-wise polynomial trajectories for controlling differential flat systems with applications to three-dimensional flight for Micro Aerial Vehicles. In this work, we propose a method to formulate trajectory generation as a quadratic program (QP) using the concept of a Safe Flight Corridor (SFC). The SFC is a collection of convex overlapping polyhedra that models free space and provides a connected path from the robot to the goal position. We derive an efficient convex decomposition method that builds the SFC from a piece-wise linear skeleton obtained using a fast graph search technique. The SFC provides a set of linear inequality constraints in the QP allowing real-time motion planning. Because the range and field of view of the robot's sensors are limited, we develop a framework of Receding Horizon Planning , which plans trajectories within a finite footprint in the local map, continuously updating the trajectory through a re-planning process. The re-planning process takes between 50 to 300 ms for a large and cluttered map. We show the feasibility of our approach, its completeness and performance, with applications to high-speed flight in both simulated and physical experiments using quadrotors.

Robust Stereo Visual Inertial Odometry for Fast Autonomous Flight

Abstract: In recent years, vision-aided inertial odometry for state estimation has matured significantly. However, we still encounter challenges in terms of improving the computational efficiency and robustness of the underlying algorithms for applications in autonomous flight with microaerial vehicles, in which it is difficult to use high-quality sensors and powerful processors because of constraints on size and weight. In this letter, we present a filter-based stereo visual inertial odometry that uses the multistate constraint Kalman filter. Previous work on the stereo visual inertial odometry has resulted in solutions that are computationally expensive. We demonstrate that our stereo multistate constraint Kalman filter (S-MSCKF) is comparable to state-of-the-art monocular solutions in terms of computational cost, while providing significantly greater robustness. We evaluate our S-MSCKF algorithm and compare it with state-of-the-art methods including OKVIS, ROVIO, and VINS-MONO on both the EuRoC dataset and our own experimental datasets demonstrating fast autonomous flight with a maximum speed of 17.5 m/s in indoor and outdoor environments. Our implementation of the S-MSCKF is available at https://github.com/KumarRobotics/msckf_vio.

Fast, autonomous flight in GPS-denied and cluttered environments

Abstract: Author(s): Mohta, K.; Mulgaonkar, Y.; Watterson, M.; Liu, S.; Qu, C.; Makineni, A.; Saulnier, K.; Sun, K.; Zhu, A.; Delmerico, J.; Karydis, K.; Atanasov, N.; Loianno, G.; Scaramuzza, D.; Daniilidis, K.; Taylor, C. J.; Kumar, V.

High speed navigation for quadrotors with limited onboard sensing

Abstract: We address the problem of high speed autonomous navigation of quadrotor micro aerial vehicles with limited onboard sensing and computation. In particular, we propose a dual range planning horizon method to safely and quickly navigate quadrotors to specified goal locations in previously unknown and unstructured environments. In each planning epoch, a short-range planner uses a local map to generate a new trajectory. At the same time, a safe stopping policy is found. This allows the robot to come to an emergency halt when necessary. Our algorithm guarantees collision avoidance and demonstrates important advances in real-time planning. First, our novel short range planning method allows us to generate and re-plan trajectories that are dynamically feasible, comply with state and input constraints, and avoid obstacles in real-time. Further, previous planning algorithms abstract away the obstacle detection problem by assuming the instantaneous availability of geometric information about the environment. In contrast, our method addresses the challenge of using the raw sensor data to form a map and navigate in real-time. Finally, in addition to simulation examples, we provide physical experiments that demonstrate the entire algorithmic pipeline from obstacle detection to trajectory execution.

Fast, Autonomous Flight in GPS-Denied and Cluttered Environments

Abstract: One of the most challenging tasks for a flying robot is to autonomously navigate between target locations quickly and reliably while avoiding obstacles in its path, and with little to no a-priori knowledge of the operating environment. This challenge is addressed in the present paper. We describe the system design and software architecture of our proposed solution, and showcase how all the distinct components can be integrated to enable smooth robot operation. We provide critical insight on hardware and software component selection and development, and present results from extensive experimental testing in real-world warehouse environments. Experimental testing reveals that our proposed solution can deliver fast and robust aerial robot autonomous navigation in cluttered, GPS-denied environments.

Planning Dynamically Feasible Trajectories for Quadrotors Using Safe Flight Corridors in 3-D Complex Environments

Abstract: There is extensive literature on using convex optimization to derive piece-wise polynomial trajectories for controlling differential flat systems with applications to three-dimensional flight for Micro Aerial Vehicles. In this work, we propose a method to formulate trajectory generation as a quadratic program (QP) using the concept of a Safe Flight Corridor (SFC). The SFC is a collection of convex overlapping polyhedra that models free space and provides a connected path from the robot to the goal position. We derive an efficient convex decomposition method that builds the SFC from a piece-wise linear skeleton obtained using a fast graph search technique. The SFC provides a set of linear inequality constraints in the QP allowing real-time motion planning. Because the range and field of view of the robot's sensors are limited, we develop a framework of Receding Horizon Planning , which plans trajectories within a finite footprint in the local map, continuously updating the trajectory through a re-planning process. The re-planning process takes between 50 to 300 ms for a large and cluttered map. We show the feasibility of our approach, its completeness and performance, with applications to high-speed flight in both simulated and physical experiments using quadrotors.

Uncertainty-Aware Reinforcement Learning for Collision Avoidance

Abstract: Reinforcement learning can enable complex, adaptive behavior to be learned automatically for autonomous robotic platforms. However, practical deployment of reinforcement learning methods must contend with the fact that the training process itself can be unsafe for the robot. In this paper, we consider the specific case of a mobile robot learning to navigate an a priori unknown environment while avoiding collisions. In order to learn collision avoidance, the robot must experience collisions at training time. However, high-speed collisions, even at training time, could damage the robot. A successful learning method must therefore proceed cautiously, experiencing only low-speed collisions until it gains confidence. To this end, we present an uncertainty-aware model-based learning algorithm that estimates the probability of collision together with a statistical estimate of uncertainty. By formulating an uncertainty-dependent cost function, we show that the algorithm naturally chooses to proceed cautiously in unfamiliar environments, and increases the velocity of the robot in settings where it has high confidence. Our predictive model is based on bootstrapped neural networks using dropout, allowing it to process raw sensory inputs from high-bandwidth sensors such as cameras. Our experimental evaluation demonstrates that our method effectively minimizes dangerous collisions at training time in an obstacle avoidance task for a simulated and real-world quadrotor, and a real-world RC car. Videos of the experiments can be found at this https URL.

OpenVINS: A Research Platform for Visual-Inertial Estimation

Abstract: In this paper, we present an open platform, termed OpenVINS, for visual-inertial estimation research for both the academic community and practitioners from industry. The open sourced codebase provides a foundation for researchers and engineers to quickly start developing new capabilities for their visual-inertial systems. This codebase has out of the box support for commonly desired visual-inertial estimation features, which include: (i) on-manifold sliding window Kalman filter, (ii) online camera intrinsic and extrinsic calibration, (iii) camera to inertial sensor time offset calibration, (iv) SLAM landmarks with different representations and consistent First-Estimates Jacobian (FEJ) treatments, (v) modular type system for state management, (vi) extendable visual-inertial system simulator, and (vii) extensive toolbox for algorithm evaluation. Moreover, we have also focused on detailed documentation and theoretical derivations to support rapid development and research, which are greatly lacked in the current open sourced algorithms. Finally, we perform comprehensive validation of the proposed OpenVINS against state-of-the-art open sourced algorithms, showing its competing estimation performance.

Survey on Computer Vision for UAVs: Current Developments and Trends

Abstract: During last decade the scientific research on Unmanned Aerial Vehicless (UAVs) increased spectacularly and led to the design of multiple types of aerial platforms. The major challenge today is the development of autonomously operating aerial agents capable of completing missions independently of human interaction. To this extent, visual sensing techniques have been integrated in the control pipeline of the UAVs in order to enhance their navigation and guidance skills. The aim of this article is to present a comprehensive literature review on vision based applications for UAVs focusing mainly on current developments and trends. These applications are sorted in different categories according to the research topics among various research groups. More specifically vision based position-attitude control, pose estimation and mapping, obstacle detection as well as target tracking are the identified components towards autonomous agents. Aerial platforms could reach greater level of autonomy by integrating all these technologies onboard. Additionally, throughout this article the concept of fusion multiple sensors is highlighted, while an overview on the challenges addressed and future trends in autonomous agent development will be also provided.