The international Field Robot Event is an annual contest on an agricultural field, where students and their supervisors compete within several tasks in autonomous navigation and other operations. Every year different tasks need to be performed during the event. The Field Robot Event has been founded by the Wageningen University in 2003, in order to motivate students to develop autonomous field robots. We are looking forward to the 21th event and hope to enjoy creative and functional solutions. The agricultural tasks will be challenging for the robots and their students, but behind engineering skills the organisation wants to promote meeting international colleagues ‐ and of course having fun during the contest!

Task 1 – navigation

For this task, the robots are navigating autonomously through a real maize field. Turning must follow adjacent rows for track 1 to 5. From exiting track 5 the robot must follow a given particular turning pattern. This task is all about accuracy, smoothness, and speed of the navigation operation between the rows. Within three minutes the robot navigates between the rows. The aim is to cover as much travelled distance as possible. You find an example field and driving pattern in Figure 1.1. The first 3 tracks are without intra-row gaps to make it easy for robots to start. The rest of the field – track 4 to 11 – there are intra-row gaps even sometimes on both sides. In the last part – after track 5 – the robot has to follow a particular given turning and row pattern. The pattern may look as: S – 1L – 1R – 3L – 2L – 2R – F.

Random stones and pebbles are placed along the path. Therefore, machine ground clearance is required. In order to make it easier for sensors there will be no gaps at the row entries and exits. The ends or beginnings of the rows may not be in the same line. The headland will be perhaps indicated by a fence or ditch or similar.

Task 2 – autonomous counting of maize plants

Task 2 is an extension of task 1. In addition to the first task of autonomous navigation in a maize field, the field robots must count the number of plants in each row and present the results row by row immediately after finishing the individual rows by providing the data using a REST API. So, the field robots are required to continually provide data while in the field. And after finishing the task the robots must also immediately provide a JSON file on a USB stick of the number of counted maize plants. 

TASK 3 – mapping of weeds in a grassland area

In this task, the field robots must work in a grassland area with the dimensions of approximately 8m x 8m, see figure below. The task is to locate and map any weeds in the area. To support the localisation within the grassland area, it will be enclosed by a side fence, e.q. straw bales or something similar with a height of at least 0.3 meters. The result data consisting of (x, y) data sets must be provided by the field robot by using the above-mentioned REST interface. Additional obstacles might be present in the grassland area and must be avoided by the robot. The weeds will be represented by small artificial flowers placed on the ground. 

TASK 4 – Weed control in a Grassland Area

In Task 4, field robots receive weed locations at the start via REST and must navigate to these spots for weed control within 1-3 minutes. No predefined track exists, and robots must autonomously move to the locations. Weed control methods are unspecified but must be verifiable, so spot spraying, removing the plant with a robotic arm or burning the weeds are all valid options however merely driving over weeds does not earn points. The objective is to effectively apply weed control measures. Scoring involves two aspects: distance points for reaching weeds within specific proximity (1 point for <0.5 meters, 2 points for <0.15 meters), and application points for the effectiveness of weed control. However, the details of how scoring is calculated are not provided within the summarized text. The total score is determined by the sum of individual scores for each weed.

Task 5 – Freestyle

Teams are invited to let their real robot perform freestyle on the event venue. The explanation as well as the performance must be shown to the jury and the spectators. The team must explain the idea and the machine. Comments during the robot’s performance are also welcome. Creativity and fun are required for this task as well as an application-oriented performance. The freestyle task should be related to an agricultural application. Teams will have a time limit of five minutes for the presentation including the robot’s performance.