A multi agent system for precision agriculture

Abstract

This chapter investigates the use of a multi agent system for precision agriculture. Precision agriculture refers to the management of farm operations based on observation, measurement and fast response to inter- and intra-field variability in crops. Important characteristics in this application are path following, disturbance rejection and obstacle avoidance of which path following is addressed in this chapter. This study combines the degree of freedom of aerial vehicles with the location precision of ground vehicles in order to achieve a common goal. The multi agent system in this study consists of a quadrotor as the aerial agent and tracked robots with a rotary camera as the agents on the ground to achieve the common task of following a predefined path while maintaining the formation. This research uses a combination of low-level PID cascade control for the ground vehicles with a high level predictive control for the aerial agent to ensure optimal control of the positions of the ground robots and the quadrotors. A series of proof-of-concept experiments for this novel combined control strategy are performed. Simulation and experimental results suggest that the proposed control system is able to maintain the formation of the ground vehicles and provide a good tracking of the ground vehicles by the quadrotor. Because of the setup of the described system, it has also potential applications in traffic control and analysis, search and rescue operations, etc.