An effective collaboration evolutionary algorithm for multi-robot task allocation and scheduling in a smart farm

Abstract

The rapid development of intelligent and unmanned technologies has created new opportunities and challenges for smart farms. Agricultural production is increasingly incorporating intelligent robots for various farming activities. This paper investigates the task allocation and scheduling for multiple agricultural robots operating in a smart farm. For addressing this problem, a collaborative discrete bee colony algorithm (CDABC) is introduced to provide a viable solution, aiming to optimize the system efficiency in smart farms. To enhance the global search ability, several solution neighborhoods are designed based on a single robot and multi robots, and a problem-oriented dynamic neighborhood strategy list is given for the collaborative employment bees stage. Additionally, the neighborhoods based on the critical robot are constructed to enhance the local exploitation capability. A novel collaborative following bee strategy is also integrated to help individuals trapped in the local optimum move to more ideal areas in the scout bee stage. Finally, comprehensive design experiments are carried out to determine optimal parameter configurations and demonstrate that CDABC has the better performance than the state-of-the-art algorithms.