A long-cherished vision of drones is to autonomously traverse through clutter to reach every corner of the world using onboard sensing and computation. In this paper, we combine onboard 3D lidar sensing and sim-to-real reinforcement learning (RL) to enable autonomous flight in cluttered environments. Compared to vision sensors, lidars appear to be more straightforward and accurate for geometric modeling of surroundings, which is one of the most important cues for successful obstacle avoidance. On the other hand, sim-to-real RL approach facilitates the realization of low-latency control, without the hierarchy of trajectory generation and tracking. We demonstrate that, with design choices of practical significance, we can effectively combine the advantages of 3D lidar sensing and RL to control a quadrotor through a low-level control interface at 50Hz. The key to successfully learn the policy in a lightweight way lies in a specialized surrogate of the lidar’s raw point clouds, which simplifies learning while retaining a fine-grained perception to detect narrow free space and thin obstacles. Simulation statistics demonstrate the advantages of the proposed system over alternatives, such as performing easier maneuvers and higher success rates at different speed constraints. With lightweight simulation techniques, the policy trained in the simulator can control a physical quadrotor, where the system can dodge thin obstacles and safely traverse randomly distributed obstacles.

Due to the limited online computational resources and the inherent probability of hardware and software failures of real-world robots, large-scale formation planning faces two common challenges: computational intractability and agent failures. Based on the theory of sparse graphs and the maximum clique, we achieve a resilient and efficient formation planning (RE-Formation) to address these issues. To improve the computational efficiency of trajectory planning while ensuring flexible formation maneuvers, we introduce sparse graphs to describe connection relationships and present a sparse graph construction method with closed-form solutions. The sparse graphs ensure theGlobalRigidity for uniquely corresponding to a geometric shape andPreserve the mainFeatures of complete graphs, denoted as the GRPF sparse graph. To prevent the impact of abnormal agents, the problem of eliminating abnormal agents is transformed into an outlier rejection problem that can be solved by computing the maximum clique. We approximate the maximum clique by periodically triggering the calculation of the maximumk-coreto meet the real-time computational demands of large-scale swarms. We validate the performance through real-world experiments and implement formation planning with 100 drones in simulation. Benchmark comparisons and ablation experiments demonstrate the effectiveness of our method.

Drone racing has become a popular international competition and has attained wide attention in recent years. However, the requirements of high-level operation keep the novice pilots away from participating in it. This paper presents a trajectory-based flight assistive system that enables various operators to fly the drone in a racing scene at a high speed. The whole system is structured hierarchically, consisting of both offline and online components. In the offline part, a global time-optimal trajectory is generated as the expert reference, and a dense flight corridor is constructed to provide sufficiently large safe region. In the online part, a remote control-mapped primitive is designed to fast encapsulate pilots’ inputs, and the time mapping based trajectory progress is customized to further capture intention. Then, a trajectory planner is proposed to generate intention-aligned, smooth, feasible, and safe trajectories periodically. Additionally, a yaw planning that provides the pilot with the best suitable view angle is employed to further alleviate the operation difficulty. Simulations and real world experiments are implemented to verify the performance of our system. The maximum flight speed can reach 6.0 m/s for a novice drone pilot in a real racing scene. Our code is released as an open-source package1.

Reassignment strategies are of great significance to improve the dispatching efficiency of the ride-sourcing market by reassigning drivers and passengers. However, due to the focus on the feasibility of the reassignment strategy in the short period, previous studies ignore possible reassignment opportunities in the future and inevitably make short-sighted reassignment decisions. To fully exploit the effect of the reassignment strategy, this study proposes a two-stage reassignment framework, which integrates a reinforcement learning algorithm and the bilateral matching reassignment model. The Markov decision process is adopted to dynamically model the reassignment problem. In the framework, the reinforcement learning algorithm is utilized to first learn the randomness and dynamics of travel patterns from historical data and select vehicles participating in the reassignment process. Then, the bilateral matching reassignment model formulates the matching relationship after reassignment for passengers (drivers). Furthermore, for the cases where reassignment may increase individual matching distance, a personalized bilateral matching reassignment model is developed to avoid that. Experiments based on real data in Beijing found that learning passenger travel patterns and adjusting vehicle reassignment moments can greatly improve the passenger experience and reduce driving costs. The results also suggest that the efficiency of the reassignment strategy is influenced by the supply and demand conditions of the ride-sourcing system. This justifies that the framework can be applied to optimize the dispatching process, reduce carbon emissions, and build an eco-friendly travel system.