I need help for Genetic Algorithm
15 Ansichten (letzte 30 Tage)
Ältere Kommentare anzeigen
I'm using the Robotics Toolbox for Matlab to built the model of the robot and to get all kinematic and dynamics since this way is more easy than manual calculation, then I use the ode15s solver to solve the nonlinear differential equations of the robot with my controller. All thing are OK and I have a result, but I want to have a best result by choosing the best values for the controller parameters and gains, for that I'm trying to use the Genetic Algorithm but really I cant understand how to manage them together (ode15s and Genetic Algorithm, etc..), so if there is someone who know how to deal with Genetic Algorithm to tune the controller parameters and gains I will be happy to learn from him.
I have alot of variables need to tune there values.
The links bellow contains my code files and the Robotics Toolbox version which I use. Notice that these codes are without GA.
To run the codes just only run the file AFC_fivelinkmain.m. Some variables which I want to tune them are bellow:
1: k=diag([k1 k2 k3]) is a positive diagonal matrix
2: kp is positive constant
3: r0 is positive constant
4: rv0 is positive constant
Will wait your response and I hope someone will help me.
Best regards
0 Kommentare
Antworten (1)
Shashank Prasanna
am 31 Jan. 2013
[Your link is dead]
What have you tried so far using GA?
like any optimization problem if you can set up your objective function to take your tunable parameters as inputs minimizing some performance metric.
Put your ODE in a function which takes a vector of inputs for the controller parameters. Calling this function for a particular set of coefficients should solve the ODE.
You should have some metric to minimize, maybe meeting some design specifications? That is for you to decide. Create a new function that minimizes the metric by calling the above ODE function. Once you have done that call GA to minimize the function for some 'x' which is a vector of all the controller parameters.
Here is an example of optimizing an ODE using patternsearch and GA, this should be a start:
0 Kommentare
Siehe auch
Kategorien
Mehr zu Genetic Algorithm finden Sie in Help Center und File Exchange
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!