Implementation of UAV Smooth Path Planning by Improved Parallel Genetic Algorithm on Controller Area Network

Unmanned aerial vehicle (UAV) path planning is an essential branch in UAVs research. This paper presents the hardware implementation of the UAV path planning problem using an improved parallel genetic algorithm (GA) in a multi-microcontroller environment. A controller area network (CAN) bus is a robust bus designed to allow microcontrollers to communicate with each other in applications without a host computer. The CAN bus is used to communicate between the microcontrollers and solve the path planning problem with the parallel algorithm. The data exchange on this network is by the multi-master model, so it is possible to implement an asynchronous and multi-master parallel algorithm using CAN bus. Also, we use the 32-bit ARM Cortex-M3 microcontroller (with CPU clock up to 100 MHz) for hardware implementation. The comparison of both single and parallel GA shows that a multi-microcontroller structure produces better results on the CAN bus, and the parallel version experiences significantly faster speeds than the sequential version.

Automated summary

This paper presents a hardware implementation of the UAV path planning problem using an improved parallel genetic algorithm and a multi-microcontroller structure with CAN bus communication. The results show that the parallel version performs significantly faster on the CAN bus and produces better results compared to the sequential version.