search

Liyucheng1997 / Motion-planning-with-adaptive-sampling

This is all the codes in my master thesis related to automated driving.
0 stars 0 forks source link

How do "simple speed qp planning" and "ST dynamic programming" work together to achieve adaptive sampling in speed planning? #1

Open mrayur opened 1 year ago

mrayur commented 1 year ago

Hi Liu,

Thanks for your sharing about your work!

I noticed that the ST dynamic programming is commented out in the Simulink, the part linked with the main body is the "simple speed qp planning" module, I tried to commented out the "simple speed qp planning" module, and linked the "ST dynamic programming" with the main body, I found that the simulation run much slower, Q1: so the advantage taken by "ST dynamic programming" is not obvious, what's more, the "simple speed qp planning" doesn't consider the dynamic obstacles, so the comparison in the aspect of real-time performance is hard to make, right?

Q2: Can I use the output of "ST dynamic programming" (that is dp_s, dp_s_dot, dp_s_dot2, and dp_relative_time) as the input of "simple speed qp planning" module? Thus leading to a "DP"+"QP" procedure?

Thanks in advance! Looking forward to your reply!

mrayur commented 1 year ago

And I found that the CarSim S-Function is totally cut off (no input or output) in crossroad_cs.slx, so the ego car is not controlled?