Click on “Download PDF” for the PDF version or on the title for the HTML version.

If you are not an ASABE member or if your employer has not arranged for access to the full-text, Click here for options.

A research project on development of autonomous machines for agricultural operations is currently in progress. The goal has been to develop and demonstrate the benefits from using an autonomous platform and information system for crop surveillance. The unit should navigate and operate autonomously in the field while carrying out crop analysis tasks with computer-vision. A key component, which is being developed as a simulation prototype, is an integrated management system, which involves planning and activation of a job and generation of sampling points and routes.

An overall system architecture has been specified in the object-oriented Unified Modeling Language (UML) and is used to develop a computerized simulation of the operation of the system. As the basis for the planning, the system interfaces to a prototype farm management information system. The planning domain of jobs in the field is modeled in relation to such objects as field, crop, machines, etc., and the attributes of these objects. The attributes specify the kind of data relevant to the decision-making processes involved in the planning and the implementation of the jobs.

The route planning system interfaces to the overall management system and relays concrete sequences of way-points to the vehicle-level controller. It includes a new algorithm prescribing an optimal sampling and a corresponding efficient route plan for the movements of the unit in the field. The planning algorithm has been designed on the basis of an evaluation of the spatial and temporal variability of weed occurrence and weed intensity.

The management tasks related to the use of autonomous platform units have been identified. A hierarchical management system, including specific plans for sampling points and routes aimed at weed mapping is demonstrated. Furthermore, the management system and its database provides the baseline for further enhancements, like inclusion of explicit scheduling models or specific decision support models for evaluating acquired information from the field.