.png){kind=icon}
.png){kind=icon}
Return
Optimization of task assignment for multi-farm multi-weeding robots based on discrete artificial bee colony algorithm
.png){kind=icon}
PRE
AI
DOI：10.1016/j.eswa.2024.126182.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
As smart agriculture continues to advance, the potential of agricultural robots to enhance productivity and lower costs becomes increasingly evident. This paper investigates the multi-farm multi-weeding robot task assignment problem (MFMWRTA) with minimization of the maximum completion time. A multi-farm discrete artificial bee colony (MFDABC) algorithm based on the improved discrete artificial bee colony (DABC) algorithm is presented to address this problem. The algorithm first generates high-quality initial solutions using the proposed multi-farm NEH (MFNEH) based heuristic and iterative greedy (IG) based heuristic, then assigns an appropriate number of weeding robots to each farm. During the employed bee and scout bee phases, this study develops five local search operators to enhance the algorithm's ability to perform local searches. During the onlooker bee phase, integrating the tournament selection strategy with the key robot concept assists the algorithm in escaping local optima. Experimental results show that the MFDABC algorithm outperforms existing heuristics on several test instances with better optimization performance and robustness.
Keywords:
Scheduling
Agricultural robots
Task assignment and scheduling
Heuristics
Multi-farm
Discrete artificial bee colony algorithm
Journal
IF:
7.5
Papers:
2.9W
Citations:
10.2W