Multi-agent planning has numerous real-world applications, such as autonomous agriculture, search and rescue, and infrastructure inspection. It is thus of tremendous benefit to improve on the state-of-the-art in multiple metrics to make autonomous multi-agent systems more efficient and a more viable real-world solution. In this letter, we propose a new framework for multi-agent planning in a static environment that improves upon the existing state-of-the-art in multiple areas such as computation time, trajectory smoothness, and feasibility. The proposed method first generates a global path that only avoids static obstacles and then generates a Safe Corridor around it. It then extends the notion of Safe Corridors and makes them time-aware in order to account for the future positions of other agents. The time-aware Safe Corridor, along with a local reference trajectory generated from the global path, are used in an Mixed-Integer Quadratic Program/Model Predictive Control formulation that generates a collision-free optimal trajectory. The proposed framework is real-time, decentralized, and synchronous. It is compared to two recent state-of-the-art methods in simulations. It outperforms both methods in robustness as well as feasibility and computation time.

Coordinated flight of multiple drones allows to achieve tasks faster such as search and rescue and infrastructure inspection. Thus, pushing the state-of-the-art of aerial swarms in navigation speed and robustness is of tremendous benefit. In particular, being able to account for unexplored/unknown environments when planning trajectories allows for safer flight. In this work, we propose the first high-speed, decentralized, and synchronous motion planning framework (HDSM) for an aerial swarm that explicitly takes into account the unknown/undiscovered parts of the environment. The proposed approach generates an optimized trajectory for each planning agent that avoids obstacles and other planning agents while moving and exploring the environment. The only global information that each agent has is the target location. The generated trajectory is high-speed, safe from unexplored spaces, and brings the agent closer to its goal. The proposed method outperforms four recent state-of-the-art methods in success rate (100% success in reaching the target location), flight speed (97% faster), and flight time (50% lower). Finally, the method is validated on a set of Crazyflie nano-drones as a proof of concept.

Aerial swarms can substantially improve the effectiveness of drones in applications such as inspection, monitoring, and search for rescue. This is especially true when those swarms are made of several individual drones that use local sensing and coordination rules to achieve collective motion. Despite recent progress in swarm autonomy, human control and decision-making are still critical for missions where lives are at risk or human cognitive skills are required. However, first-person-view (FPV) teleoperation systems require one or more human operators per drone, limiting the scalability of these systems to swarms. This work investigates the performance, preference, and behaviour of pilots using different FPV interfaces for teleoperation of aerial swarms. Interfaces with single and multiple perspectives were experimentally studied with humans piloting a simulated aerial swarm through an obstacle course. Participants were found to prefer and perform better with views from the back of the swarm, while views from the front caused users to fly faster but resulted in more crashes. Presenting users with multiple views at once resulted in a slower completion time, and users were found to focus on the largest view, regardless of its perspective within the swarm.

Planification de mouvement mono et multi-agents pour multirotors à haute vitesse: La navigation autonome des drones aériens a de nombreuses applications réelles qui peuvent rendre certaines tâches plus rapides et plus efficaces, telles que la recherche et le sauvetage. L'approche principale consiste à diviser le problème de la navigation autonome en sous-problèmes et à essayer de les résoudre de manière optimale. Ces sous-problèmes sont généralement considérés comme étant la perception (localisation et cartographie), la planification et le contrôle. Dans ce travail, nous abordons certains des sous-problèmes qui constituent des goulots d'étranglement du vol rapide et agile de la navigation autonome par drone. Nous mettons l'accent sur l'adaptation de nos algorithmes aux systèmes embarqués à faible puissance de calcul. Notre travail se scinde en 4 parties. La première partie concerne un nouvel algorithme de planification hors ligne, en environnment cartographié et statique, qui bat toutes les méthodes de l'état de l'art en termes de génération de trajectoire optimale en temps pour les multirotors. La deuxième partie traite de la cartographie et étudie les limites de l'utilisation d'un GPU pour transformer une carte de nuages de points générés par des capteurs en une grille de voxels. L'accent est mis sur la génération de la grille de voxels dans le temps de calcul le plus court possible pour la rendre adaptée aux systèmes embarqués à faible puissance de calcul. La troisième partie aborde, en partant d'une grille de voxel, le problème de la génération de couloirs sûrs qui sont utilisés dans les méthodes de planification de l'état de l'art pour planifier des trajectoires sûres et réalisables. Dans notre travail sur les couloirs sûrs, nous améliorons l'état de l'art en termes de sécurité, tout en restant dans les contraintes strictes des systèmes à faible puissance de calcul. La quatrième et dernière partie utilise nos travaux sur les couloirs sûrs et propose un nouveau cadre de planification améliorant l'état de l'art de la planification multirotor dans un environnement statique/dynamique pour la planification mono/multi-agent.