An Evolutionary Approach To Robot Scheduling In Protected Cultivation Systems For Uninterrupted And Maximization Of Working Time

The protected cultivation system, an alternative to open field cultivation provides opportunities such as yearround crop production and improved food security especially during disasters as well as ease in automation. However, protected cultivation is limited by the hazardous work environments and skilled labor shortages thus necessitating robotic applications. Robots are mostly battery-powered, requiring regular charges depending on the task. In a multi-robot system, due to the limitation on the availability of charging infrastructure and uneven discharge rates of the robots depending on the task, it is very difficult to predict when the robots would require charging. Therefore, to maximize the continuous work time of the robots, optimal scheduling is required. Consequently, we propose a novel system for efficiently utilizing mobile robotic systems in protected cultivation by developing a scheduling system that maximizes work time and minimizes concentrated energy demand. We formulated the robot scheduling problem to evaluate battery charge state regularly and optimally send the robot to the charging station. This problem was solved using an evolutionary algorithm. We considered: a) the number of available robots; b) number of charging stations; c) required work hours; d) robot battery capacity; e) robot battery charge and discharge rates; and f) the number of continuous discharge time instances. All parameters could be set to user preference. The applicability of the proposed method was demonstrated with experimental simulations using MATLAB under different cases and scenarios. These cases and scenarios demonstrated that our proposed system maximized worktime by a significant percentage and minimized the required power to charge the batteries in all situations.