Robotics in agriculture has experienced an enormous development in the past decade. In order to accomplish numerous agricultural tasks, the order of which the crop is covered is really important to minimize the travelling cost and to preserve the soil conditions. In agricultural environment, several path alternatives are available for the mobile robot to travel between the crops. However, it is hard to determine an optimized sequential path for the robot to minimize cost, while ensuring a fully completed agricultural operation. This paper presents an implementation of multi-objective algorithm to solve the path planning problem in performing the fungicide spraying operation inside the greenhouse. The spraying operation problem found in the greenhouse has been expressed in terms of the travelling salesman problem found in operations research where the selected crops represent the cities and the mobile robot represents the salesman. Routes between the crops are generated using probabilistic roadmap (PRM) path planner based on the designed virtual environment. For headland turning, U-turn and π- turn has been considered to be implemented in this work. The virtual environment has been designed based on the real greenhouse environment to visualize the agricultural operation inside the greenhouse. To determine the best routes for the mobile robot, Multi-Objective Genetics Algorithm (MOGA) and Multi-Objective Particle Swarm Optimization algorithm (MOPSO) are applied and compared. In Objectives evaluation, path length and degree of rotation are used to generate the fitness function for the path. In addition, constraint such as the spraying capacity is also considered in this work. Simulation results in virtual environment are executed via MATLAB and the performance of the algorithm is compared based on execution time and path quality. The path quality assessment is conducted based on travel time, distance travelled and controller error. The best path has been chosen to be applied in executing spraying tasks in the greenhouse environment.
