The use of aerial swarms to solve real-world problems has been increasing steadily, accompanied by falling prices and improving performance of communication, sensing, and processing hardware. The commoditization of hardware has reduced unit costs, thereby lowering the barriers to entry to the field of aerial swarm robotics. A key enabling technology for swarms is the family of algorithms that allow the individual members of the swarm to communicate and allocate tasks amongst themselves, plan their trajectories, and coordinate their flight in such a way that the overall objectives of the swarm are achieved efficiently. These algorithms, often organized in a hierarchical fashion, endow the swarm with autonomy at every level, and the role of a human operator can be reduced, in principle, to interactions at a higher level without direct intervention. This technology depends on the clever and innovative application of theoretical tools from control and estimation. This paper reviews the state of the art of these theoretical tools, specifically focusing on how they have been developed for, and applied to, aerial swarms. Aerial swarms differ from swarms of ground-based vehicles in two respects: they operate in a three-dimensional space and the dynamics of individual vehicles adds an extra layer of complexity. We review dynamic modeling and conditions for stability and controllability that are essential in order to achieve cooperative flight and distributed sensing. The main sections of this paper focus on major results covering trajectory generation, task allocation, adversarial control, distributed sensing, monitoring, and mapping. Wherever possible, we indicate how the physics and subsystem technologies of aerial robots are brought to bear on these individual areas.

This article proposes a novel learning-based control policy with strong generalizability to new environments that enables a mobile robot to navigate autonomously through spaces filled with both static obstacles and dense crowds of pedestrians. The policy uses a unique combination of input data to generate the desired steering angle and forward velocity: a short history of lidar data, kinematic data about nearby pedestrians, and a subgoal point. The policy is trained in a reinforcement learning setting using a reward function that contains a novel term based on velocity obstacles to guide the robot to actively avoid pedestrians and move toward the goal. Through a series of 3-D simulated experiments with up to 55 pedestrians, this control policy is able to achieve a better balance between collision avoidance and speed (i.e., higher success rate and faster average speed) than state-of-the-art model-based and learning-based policies, and it also generalizes better to different crowd sizes and unseen environments. An extensive series of hardware experiments demonstrate the ability of this policy to directly work in different real-world environments with different crowd sizes with zero retraining. Furthermore, a series of simulated and hardware experiments show that the control policy also works in highly constrained static environments on a different robot platform without any additional training. Lastly, several important lessons that can be applied to other robot learning systems are summarized.

Target tracking is a fundamental problem in robotics research and has been the subject of detailed studies over the years. In this paper, we generate a data-driven target model from a real-world dataset of taxi motions. This model includes target motion, appearance, and disappearance from the search area. Using this target model, we introduce a new formulation of the mobile target tracking problem based on the mathematical concept of random finite sets. This formulation allows for tracking an unknown and dynamic number of mobile targets with a team of robots. We show how to employ the probability hypothesis density filter to simultaneously estimate the number of targets and their positions. Next, we present a greedy algorithm for assigning trajectories to the robots to allow them to actively track the targets. We prove that the greedy algorithm is a two-approximation for maximizing submodular tracking objective functions. We examine two such functions: the mutual information between the estimated target positions and future measurements from the robots and a new objective that maximizes the expected number of targets detected by the robot team. We provide extensive simulation evaluations to validate the performance of our data-driven motion model and to compare the behavior and tracking performance of robots using our objective functions.

We present Splat-Nav, a real-time robot navigation pipeline for Gaussian Splatting (GSplat) scenes, a powerful new 3D scene representation. Splat-Nav consists of two components: 1) Splat-Plan, a safe planning module, and 2) Splat-Loc, a robust vision-based pose estimation module. Splat-Plan builds a safe-by-construction polytope corridor through the map based on mathematically rigorous collision constraints and then constructs a Bézier curve trajectory through this corridor. Splat-Loc provides real-time recursive state estimates given only an RGB feed from an on-board camera, leveraging the point-cloud representation inherent in GSplat scenes. Working together, these modules give robots the ability to recursively re-plan smooth and safe trajectories to goal locations. Goals can be specified with position coordinates, or with language commands by using a semantic GSplat. We demonstrate improved safety compared to point cloud-based methods in extensive simulation experiments. In a total of 126 hardware flights, we demonstrate equivalent safety and speed compared to motion capture and visual odometry, but without a manual frame alignment required by those methods. We show online re-planning at more than 2 Hz and pose estimation at about 25 Hz, an order of magnitude faster than Neural Radiance Field (NeRF)-based navigation methods, thereby enabling real-time navigation. We provide experiment videos on our project page at https://chengine.github.io/splatnav/. Our codebase and ROS nodes can be found at https://github.com/chengine/splatnav.

In this paper we examine the problem of autonomously exploring and mapping an environment using a mobile robot. The robot uses a graph-based SLAM system to perform mapping and represents the map as an occupancy grid. In this setting, the robot must trade-off between exploring new area to complete the task and exploiting the existing information to maintain good localization. Selecting actions that decrease the map uncertainty while not significantly increasing the robot's localization uncertainty is challenging. We present a novel information-theoretic utility function that uses both Shannon's and Rényi's definitions of entropy to jointly consider the uncertainty of the robot and the map. This allows us to fuse both uncertainties without the use of manual tuning. We present simulations and experiments comparing the proposed utility function to state-of-the-art utility functions, which only use Shannon's entropy. We show that by using the proposed utility function, the robot and map uncertainties are smaller than using other existing methods.