Cooperative autonomous search, grasping, and delivering in a treasure hunt scenario by a team of unmanned aerial vehicles

Abstract

This paper addresses the problem of autonomous cooperative localization, grasping and delivering of colored ferrous objects by a team of unmanned aerial vehicles (UAVs). In the proposed scenario, a team of UAVs is required to maximize the reward by collecting colored objects and delivering them to a predefined location. This task consists of several subtasks such as cooperative coverage path planning, object detection and state estimation, UAV self‐localization, precise motion control, trajectory tracking, aerial grasping and dropping, and decentralized team coordination. The failure recovery and synchronization job manager is used to integrate all the presented subtasks together and also to decrease the vulnerability to individual subtask failures in real‐world conditions. The whole system was developed for the Mohamed Bin Zayed International Robotics Challenge (MBZIRC) 2017, where it achieved the highest score and won Challenge No. 3—Treasure Hunt. This paper does not only contain results from the MBZIRC 2017 competition but it also evaluates the system performance in simulations and field tests that were conducted throughout the year‐long development and preparations for the competition.

Unsupervised learning‐based flexible framework for surveillance planning with aerial vehicles

“…The reactive collision avoidance based on the MPC predictions uses a slight alteration of the desired trajectory altitude if the MPC predictions contain collisions between the vehicles. The used MPCbased collision avoidance is partially described in Spurný et al (2018) and it is presented in Báča et al (2018). Besides, the found solutions and especially those found by the proposed unsupervised learning are such that the found trajectories are mutually noncrossing, and thus collision-free (see a discussion on that in Faigl (2016).…”

“…Thus, the problem is to determine a sequence of visits to the object locations for each vehicle together with the corresponding trajectories connecting the waypoints from which objects are captured such that all the objects are identified as quickly as possible, and the vehicles return to their initial locations. The expected computational requirements for the surveillance planning and the specific setup of the MBZIRC 2017 deployment allow to relax the collision avoidance in the planning part, and it is addressed by the reactive collision avoidance implemented in the used MPCbased trajectory following controller (Báča, Hert, Loianno, Saska, & Kumar, 2018;Spurný et al, 2018). Therefore, an explicit finding of collision-free trajectories is not considered in the following problem formulations.…”

“…Therefore, it has been requested to develop a surveillance planning algorithm with low computational requirements while still be able to provide solutions of satisfiable quality. Thus, our initial intention was to provide a cost-efficient solution in less than 1 s using a single core of a conventional computer with a central processing unit (CPU) of the iCore7 class running at the frequency around 3.4 GHz, that is, computational resources available at our UAVs (Spurný et al, 2018).…”

UAV Payload Transportation via RTDP Based Optimized Velocity Profiles

“…The RTDP algorithm receives the current state of the UAV and provides a velocity decision. This velocity decision can be implemented using the controller presented in this paper or more calibrated ones such as [33]. We assume that the UAV is capable of lifting a cube with a frontal area of 0.12 m 2 .…”

Full-Pose Manipulation Control of a Cable-Suspended Load With Multiple UAVs Under Uncertainties

“…Unmanned Aerial Vehicles (UAVs) are nowadays used in different domains, from search & rescue and inspection to precise agriculture and transportation of loads [1]. In the latter case, the object is either rigidly attached to the vehicle [2] or attached using a cable [3]. Multiple aerial vehicles rigidly attached to the load [4] or attached to the load's center of mass (CoM) using cables [5] are typically used to overcome limited payload of the single-vehicle solution.…”