Task allocation alogrithms for multiple UAVs of SEAD missions and VRP.
The repository is developed to tackle the dynamic task allocation problem in the Suppression of Enemy Air Defenses missions (SEAD) for heterogeneous multi-UAV systems. To perform the SEAD mission in dynamic environments, a method for decentralized dynamic task allocation based on a Decentralized Parallel Genetic Algorithm (DPGA) is employed. The method parallelizes a genetic algorithm across the UAV swarm and periodically achieves information exchange through UAV-to-UAV (U2U) communication for conflict resolution and further optimization. Based on the received information and the strategies for dynamic task allocation, each UAV can generate appropriate solutions according to the current environment and be able to confront situations involving ad-hoc addition of targets and UAV failures. Moreover, a path-following method is employed to control them to execute the assigned tasks.The ultimate objective of the repository is to implement real flight operations in outdoor environments. Validation of the developed system is achieved through the repositories located at Multi-UAV_System_UAVprogram and GroundControlStation_of_Multi-UAV_Systems .
Due to the dubins package, the program is only worked on Linux.
There are two ways to operate the repository on Windows:
Clone this repo
git clone https://github.com/jerryfungi/Multi-UAV_Task_Allocation_SEADmission.git
Install the necessary dependencies by running.
pip install -r requirements.txt
Execute the python scripts
python3 decentralized_GA_SEAD.py
python3 GA_SEAD_process.py
python3 GA_VRP.py
python3 PSO_VRP.py
Considering that the proposed thesis has not been published yet, the related research is presented below for reference.