This repo contains functions and types that are used by both the Barc simulation framework and the "real" ROS framework. It contains basic BARC-specific types (related to the model and to MPC functions) and functions for their initialization. There are also functions to simulate the car's dynamics and to control the car along a racetrack (using Learning Model Predictive Control). The library should be structured in folders according to its purposes.
- Introduced general functions and classes that can be used in simulation and on the BARC
- Added specific functions for Learning Model Predictive Control (LMPC) and system identification
- Added the option for steering delay (in the MPC formulation, approx. 2-step-delay on the BARC system (~0.2 seconds))
- Added branch for system ID with a different sampling frequency than the MPC formulation (50 Hz sysID, 10 Hz MPC)