Minimum-Time Quadrotor Waypoint Flight in Cluttered Environments

Abstract

We tackle the problem of planning a minimum-time trajectory for a quadrotor over a sequence of specified waypoints in the presence of obstacles while exploiting the full quadrotor dynamics. This problem is crucial for autonomous search and rescue and drone racing scenarios but was, so far, unaddressed by the robotics community in its entirety due to the challenges of minimizing time in the presence of the non-convex constraints posed by collision avoidance. Early works relied on simplified dynamics or polynomial trajectory representations that did not exploit the full actuator potential of a quadrotor and, thus, did not aim at minimizing time. We address this challenging problem by using a hierarchical, sampling-based method with an incrementally more complex quadrotor model. Our method first finds paths in different topologies to guide subsequent trajectory search for a kinodynamic point-mass model. Then, it uses an asymptotically-optimal, kinodynamic sampling-based method based on a full quadrotor model on top of the point-mass solution to find a feasible trajectory with a time-optimal objective. The proposed method is shown to outperform all related baselines in cluttered environments and is further validated in real-world flights at over 60km/h in one of the world's largest motion capture systems. We release the code open source.

A node selection algorithm to graph-based multi-waypoint optimization navigation and mapping

“…The multi-waypoint visiting sequence problem can be used to solve the transportation planning problem and Covid-19 disinfection robot path planning in hospitals, in which agents (vehicles) need to be delivered as well as the overall cost and time need to be minimized [14] . The algorithm of the improved PSO to finding multiwaypoint visiting sequence is explained in Algorithm 1.…”

“…Robot navigation is one of the fundamental components in robotic systems, which includes multi-waypoint navigation system [6][7][8] . As an increasing demand and limited onboard resources for autonomous robots, it requires the ability to visit several targets in one mission to optimize multiple objectives, including time, robot travel distance minimization, and spatial optimization [9][10][11][12][13][14][15] . For example, due to a global pandemic, the world struggled to sanitize heavily populated areas, such as airports, hospitals, and educational buildings.…”

“…For example, due to a global pandemic, the world struggled to sanitize heavily populated areas, such as airports, hospitals, and educational buildings. Autonomous robots with multi-waypoint navigation systems can effectively sanitize all targeted areas without endangering the workers [14,16] . As well as in agriculture management, multi-waypoint strategies allow the robotic system to navigate and survey multiple areas to assist in production and collection.…”

“…In order to employ robotic systems in real-world scenarios, one critical factor is to develop autonomous robot multi-waypoint navigation and mapping system [17] . In order to solve the autonomous robot navigation problem, countless algorithms have been developed, such as graph-based [18,19] , ant colony optimization (ACO) [20][21][22] , bat-pigeon algorithm (BPA) [23] , neural networks [24][25][26] , fuzzy logic [27] , artificial potential field (APF) [28] , sampling-based strategy [14,29] , hybrid algorithms [30] , task planning algorithm [31] , etc. Chen et al produced a hybrid graph-based reinforcement learning architecture to develop a method for robot navigation in crowds [18] .…”

“…Jensen-Nau et al [28] integrated a Voronoi-based path generation algorithm and an artificial potential field path planning method, in which the latter is capable of establishing a path in an unknown environment in real-time for robot path planning and obstacle avoidance. Penicka and Scaramuzza [14] developed a sampling-based multi-waypoint minimum-time path planning model that allows obstacle avoidance in cluttered environments. Ortiz and Yu [30] proposed a sliding control method in combination with simultaneous localization and mapping (SLAM) method to overcome the bounded uncertainties problem, which utilizes a genetic algorithm to improve path planning capabilities.…”