search

helgeanl / GP-MPC

MPC with Gaussian Process
MIT License
193 stars 60 forks source link

f_hybrid option #7

Open jwangjie opened 4 years ago

jwangjie commented 4 years ago

Hi Helge-Andre,

Thank you very much for sharing your framework, it's very inspiring.

I am particular to try your hybrid implementation where GP estimating modeling errors. I noticed in your code , you said f_hybrid option is not finished implemented. I was wondering what do you mean here? Can I still use your framework to reproduce Hewing2017? Thanks.

Regards, Jie

helgeanl commented 4 years ago

Hi, Unfortunately, I don't remember the state of that function as it has been a while since I worked on this code. I didn't use the error model implementation in the results of my thesis, but it shouldn't be too much work to get that working. It would mostly involve cleaning up the covariance equation.

qq2217942597 commented 3 years ago

Hi Helge-Andre, Thank you very much for sharing your framework, it's very inspiring. I am particular to try your hybrid GP (GP model for dynamic equations, and RK4 for kinematic equations). I noticed in your code , you said "Missing kinematic states" in the step of Hybrid output covariance matrix. Will this affect the results?

qq2217942597 commented 3 years ago

Hi Helge-Andre, Thank you very much for sharing your framework, it's very inspiring. For the F_hybrid option，do I need to make changes elsewhere while cleaning up the covariance equation? For example, "Linearize around operating point and calculate LQR gain matrix" (mpc_class.py Line 594) or "select the chosen integrator" (mpc_class.py line383)?