Average Errorpoints

Tuning the error points is very difficult, the standard number was a compromise between removing false detentions and seeing robots early enough.

The best way to improve the detection would really be to replace the error points method with something else a bit more reliable, such as matching a robot shape in the image.

If you do want to try and improve the existing detection there are a few things you can try:

Changing overtakeThreshold to tune the detection

• Larger values will reduce the number of false detections, but will reduce the likelihood of detecting robots
• Smaller values will increase the number of false detections, but will increase the likelihood of detecting robots

Changing cropY1 to see further ahead

• Smaller values will allow the robot to see further ahead and spot robots easier, but may make the basic track following harder or less reliable
• Much smaller values will allow the robot to see even further ahead and spot robots easier, but will likely cause the robot to get confused by the background and struggle to get around the track
• Larger values will cause the robot to see for shorter distances ahead and make spotting robots more difficult, but may improve the track following by removing noise
• Much larger values will cause the robot to see too short and prevent the track following from working properly

Slowing down to allow more time to detect robots

• Slowing down will give the detection a better chance of seeing robots ahead, but will cause longer lap times
• Speeding up will make detection harder, but it will also improve the lap times as long as collisions are avoided

Averaging error points over a few frames
This would need some additional code, but you could change the detection so it needs to see the detection limit for a few frames in a row.

• Averaging over more frames will remove more false detections, but will spot robots late
• Averaging over less frames will not be as much of an improvement, but will not change the detection speed as much

As all of these things affect each other you will probably need to play with the settings to see what happens. The best hing to do is test your code with some of the images taken from the MonsterBorg driving around the track and checking when the code decides their is a robot to overtake against the picture it is being shown.