Challenge 3

Challenge 3

par Paul François Christian Huot-Marchand,
Nombre de réponses : 1

Hello,

I am experiencing several issues. First, when I run the simulation with our evolved controller from Challenge 1, the simulation starts correctly, but the robot moves for about one second and then stops.

Secondly, when running the single-objective optimization using SO2, the robot barely moves.

I am not sure whether these two issues are related, but I was wondering if you might have a quick fix or suggestion.

Thank you for your help.

En réponse à Paul François Christian Huot-Marchand

Re: Challenge 3

par Fuda van Diggelen,
Hi,

1) Did you scale the legs to match the length of the challenge1, additionally did you check the termination conditions and print which one is triggered?
2) Does this occur at the beginning or does your EA find this as a best solution, if the latter: what would be the reason and how can change the environment/reward/termination conditions?

Good luck!
Fuda