An Integrated Mission Planning Framework for Sensor Allocation and Path Planning of Heterogeneous Multi-UAV Systems

Abstract

Mission planning is the guidance for a UAV team to perform missions, which plays the most critical role in military and civil applications. For complex tasks, it requires heterogeneous cooperative multi-UAVs to satisfy several mission requirements. Meanwhile, airborne sensor allocation and path planning are the critical components of heterogeneous multi-UAVs system mission planning problems, which affect the mission profit to a large extent. This paper establishes the mathematical model for the integrated sensor allocation and path planning problem to maximize the total task profit and minimize travel costs, simultaneously. We present an integrated mission planning framework based on a two-level adaptive variable neighborhood search algorithm to address the coupled problem. The first-level is devoted to planning a reasonable airborne sensor allocation plan, and the second-level aims to optimize the path of the heterogeneous multi-UAVs system. To improve the mission planning framework’s efficiency, an adaptive mechanism is presented to guide the search direction intelligently during the iterative process. Simulation results show that the effectiveness of the proposed framework. Compared to the conventional methods, the better performance of planning results is achieved.

Research on Multi-Objective Multi-Robot Task Allocation by Lin–Kernighan–Helsgaun Guided Evolutionary Algorithms

“…The start and end time of the time-window of task i are TW S i and TW E i respectively, with TW S i < TW E i . The constraints in (10) and (11) are to be handled using penalty functions, that is to say, add the below item to the objective function:…”

“…With MRS, the system's flexibility and overall robustness can be assured, and the individual robot's design can be simplified, too [8]. MRS are playing an important role in the context of Industry 4.0, applications include but are not limited to modern logistics [9], surveillance [10], intelligent manufacturing [11], wireless sensor networks [12], etc.…”

Scalable Multi-Robot Task Allocation Using Graph Deep Reinforcement Learning with Graph Normalization

“…This scenario is more realistic for mobile robotics, because mobile robots are usually compactly designed to serve a single task at a time, and in most cases, the task only needs one robot to complete it or the task can be decomposed into elements so that a single robot can pick it up. Typical applications of the ST-SR-TA problem include mobile robot swarm surveillance [5,6], factory robot automation [7,8], wireless sensor networks (WSNs) [9], transportation networks [10], etc. Tremendous efforts have been made to solve this problem in the literature, mostly attributing this problem to a vehicle routing problem (VRP) [11,12] or a multiple-traveling-salesman problem (mTSP) [7], then using heuristic or meta-heuristic methods to solve it.…”

Risks of Multipath Signal Propagation From GPS Satellites in Mountainous Areas During the Use of UAV as Supplementary Device for HEMS or SAR

“…The flights of unmanned aerial vehicles depend on the information they receive from external sources, such as remote-control information from the ground station and from the pilot-operator or autonomous flights whose trajectory and flight mission are pre-programmed in advance [1,2]. Adherence to the predefined flight trajectory is checked and corrected by sensors onboard implemented into the unmanned aerial vehicle (for example, ultrasonic or radar "scanning" of the surrounding environment) or by receiving and evaluation elements designed to receive and process information about the current position in space, from the GNSS receiving device, which enters as correction information to the built-in INS (Inertial Navigation System) unit [3,4].…”