Research on genetic algorithm optimization for agricultural machinery operation path planning
| Autor: | Song Xiuming |
|---|---|
| Jazyk: | angličtina |
| Rok vydání: | 2024 |
| Předmět: | |
| Zdroj: | Applied Mathematics and Nonlinear Sciences, Vol 9, Iss 1 (2024) |
| Druh dokumentu: | article |
| ISSN: | 2444-8656 2024-1933 |
| DOI: | 10.2478/amns-2024-1933 |
| Popis: | Food security is a critical strategic concern in agricultural production, where agrarian machinery plays a vital role as a fundamental input and a crucial tool for enhancing production efficiency. This paper details a methodology utilizing Global Navigation Satellite System (GNSS) measurement software to analyze farmland topography. This process involves terrain measurement, data acquisition concerning the terrain, and subsequent processing to create a raster map of the farmland. The path planning for agricultural machinery is then refined through an innovative application of the A* algorithm, which selects optimal routes on the raster for initial path generation. This study introduces a genetic algorithm (GA) to facilitate global path planning optimization and designs a local path planning algorithm based on a cost function. Both unobstructed and obstructed regular farmland grids are subjected to simulation analyses to evaluate the efficacy of the path-planning approach. Comparative analysis indicates that the A*-GA algorithm significantly outperforms other algorithms in various metrics, including the number of steering occurrences, the count of repeated job grids, and the job repetition rate. Specifically, when compared to the Particle Swarm Optimization (PSO) algorithm, the A*-GA algorithm demonstrates a reduction of 6.3 in the number of repeated job grids and a 2.735% decrease in the job repetition rate. Similarly, it shows a reduction of 6.2 in repeated job grids and a 2.582% decrease in the job repetition rate compared to the standalone GA algorithm. Furthermore, the enhanced genetic algorithm enables agricultural machinery to adeptly avoid obstacles, thereby ensuring operational safety and achieving the desired endpoint along the planned path. The findings underscore that the advanced genetic algorithm effectively orchestrates obstacle avoidance for agricultural robots, thus ensuring continuous operation and adherence to safety standards in agricultural machinery deployment. This integration of GNSS with advanced algorithmic strategies marks a significant advancement in precision agriculture, optimizing machinery pathways of improved farm outcomes. |
| Databáze: | Directory of Open Access Journals |
| Externí odkaz: |
|