yambati03 / dcsl-ddc Goto Github PK

scripts/optimize_tire_model.py finds the optimal parameters for the Pacejka model to minimize error against experimental data.

The validate_model.py script plots the trajectory from a given log against the trajectory predicted by the optimized dynamics model.

• The script takes in the filepath of the ROS bag to run as a command line argument.
• The predicted trajectory (green) is plotted if the -p flag is specified. It is not plotted by default.
• Plots of the heading, yaw rate, and velocity are shown prior to the simulation if the -s flag is specified.

Note: These steps assume that ROS 2 is installed.

1. Since the script loads a ROS bag that depends on custom message types (AckermannDriveStamped, ViconObject), the script must be run within a ROS workspace with the packages that define these types. First, create a ROS workspace by running mkdir -p ros_ws/src.
2. Navigate to ros_ws/src and clone this repository. Then
   • Clone the ackermann_msgs repo by running git clone -b ros2 https://github.com/ros-drivers/ackermann_msgs.git
   • Clone the vicon_ros repo by running git clone https://github.com/yambati03/vicon_ros.git
3. Build the workspace by running colcon build --symlink-install in the top-level ros_ws directory. Then, source the install directory by running . install/setup.bash
4. Then, navigate to dcsl-ddc/scripts. The validate_model.py script can be run using the following command.

python3 validate_model.py /path/to/bag.db3 -p -s