Multi-unmanned aerial vehicle autonomous navigation and task allocation algorithm for wireless self-powered communication network

Abstract

The invention discloses a multi-unmanned aerial vehicle autonomous navigation and task allocation algorithm for a wireless self-powered communication network. The algorithm comprises the steps of carrying out the combined design of user scheduling, the flight path, the flight speed and the communication mode of each unmanned aerial vehicle, and the task allocation and the path optimization between the unmanned aerial vehicles. The purposes of completing the collision-free navigation task of the multiple unmanned aerial vehicles within the specified flight time and increasing the average uplink transmission data volume of the system user to the maximum extent are achieved. An optimization problem is modeled as a Markov process, an asynchronous multi-agent deep reinforcement learning algorithm based on a shared neural network is provided, the optimization problem is solved, the neural network is trained step by step, and finally the purpose of maximizing the uplink total data volume of the system is achieved.

Description

Technical field [0001] The present invention belongs to the technical field of drone energy supply communication network, and specifically, a multi-drone autonomous navigation and task allocation algorithm for wireless self-powered communication networks. Background technique [0002] At present, with the innovation of wireless communication technology, the Internet of Things system attracted more and more research attention. On the one hand, the Internet of Things devices on the ground have information instructional requirements to achieve various infrastructed Internet services. On the other hand, the Internet of Things devices are typically affected by the energy limit. Traditionally, the wireless terminal is powered by the battery, which must be manually replaced or charge the battery to extend the service life. It usually brings high cost and inconvenience, not to be a dangerous environment (eg in a toxic environment). Therefore, the wireless transmission network (WPCN) of d...

AI Technical Summary

Problems solved by technology

[0004] However, in the current existing literature, it is usually based on the known position of the ground equipment, and the situation that the position of the UAV is unknown to the ground sensor equipment has not been considered.

And most of the existing literature is aimed at the scene of a single UAV

However, due to the limited communication range and the limited energy resources of drones, for large area scenarios, a single drone cannot always cover the entire area or keep flying for a long time

Images