[JdeRobot] Help regarding Genetic Algorithms on Gazebo

[Sakshay Mahna]

Hello!

I am Sakshay. I am currently working on the project:Robotics Academy under
Google Summer of Code. My project
<https://github.com/TheRoboticsClub/colab-gsoc2020-Sakshay_Mahna> is on
developing an exercise for students on the topic Evolutionary Robotics. The
exercise is supposed to offer a Gazebo simulation of the robot, and we have
to train the parameters of it's Neural Network by means of a Genetic
Algorithm for obstacle avoidance. However, the problem is with the
repeatability of the simulations. Every time we run the simulation, the
robot takes a slightly different course of action, which should not be the
case. The robot should behave the same way every time we run it.

It would be really great if I could get a little guidance on how to
implement this.

Thanks & Regards
Sakshay Mahna
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://gsyc.urjc.es/pipermail/jde-developers/attachments/20200812/23f21921/attachment.htm>