ROS Resources: Documentation | Support | Discussion Forum | Index | Service Status | ros @ Robotics Stack Exchange |
1 | initial version |
I don't think anyone has a dynamics model of the turtlebot 2; most projects use simple PID controllers and tune them experimentally until the provide decent results.
If you really need the system model, I suggest you gather the data and model the system yourself. In my experience, the two hardest parts of designing a good controller are capturing the system model, and doing the mathematics to verify the controller.
As an aside, I haven't heard of many people building controllers that control the torque or PWM applied to the turtlebot wheels (most people send velocities and use the built-in controller). You may want to make sure that you can actually send the control signal that you think you can send before you put a bunch of effort into building a controller that outupts torque values.
2 | No.2 Revision |
I don't think anyone has a dynamics model of the turtlebot 2; most projects use simple PID controllers and tune them experimentally until the they provide decent results.
If you really need the system model, I suggest you gather the data and model the system yourself. In my experience, the two hardest parts of designing a good controller are capturing the system model, and doing the mathematics to verify the controller.
As an aside, I haven't heard of many people building controllers that control the torque or PWM applied to the turtlebot wheels (most people send velocities and use the built-in controller). You may want to make sure that you can actually send the control signal that you think you can send before you put a bunch of effort into building a controller that outupts torque values.