There will be another testing session

Due to popular demand, we will run an additional test session with a single stationary robot in the middle of the track at the end of the first straight. This is so competitors can test real life avoidance code.

This will be at the usual time (6.30pm GMT) on the 13th December, with code due on the 11th December.

This will be in the format of the time trial - up to two minutes driving, with timers on, however if a robot does not exhibit any attempt at avoidance, we may pick them up and put out the next robot.

Please participate only if you are really testing avoidance code.

Some links to blogs that might be helpful in avoidance:

https://www.formulapi.com/blog/overtaking-tests

https://www.formulapi.com/blog/detect-traffic

https://www.formulapi.com/blog/monster-avoidance-raw-footage

https://www.formulapi.com/blog/monster-stuck-detection-fix

https://www.formulapi.com/blog/monster-raw-footage-analysis

Additional private testing sessions will be made available for testing complex code

We've decided to add a couple of private testing sessions for those who are trying out complicated code changes.

Whether you qualify or not will be entirely up to our discretion, but there will be a few general rules of thumb:

Small changes will NOT qualify. For example:

• Varying PID coefficients
• Varying lane choice
• Varying camera rate

Large changes are likely to qualify. For example:

• Changing language to something other than python
• Neural network code
• Self learning code
• Significant image processing changes

If you are doing something complex like this, please send us an email to ask if you qualify.

Thanks,

The Formula Pi bunch.