MPC for self-driving car control

Overview

This section provides a brief introduction to the project conducted as part of the Advanced Control Methods course at Skoltech in 2024. It includes the fundamental objectives of the project, information about the team members, and a link to the final presentation.

• Course: Advanced Control Methods, Skoltech, 2024
• Team Members: Zakhar Yagudin, Artem Myshlyaev, Aibek Akhmetkazy
• Final Presentation: Link

Table of Contents

• Overview
• Problem Statement
• Results
• Run the Project
• References

---

Problem Statement

The number of car crashes and accidents are rising every year. The goal of this project work is to decrease these traffic fatalities by improving tracking accuracy and ensuring safety constraints using Model Predictive Control for self-driving cars.

---

Results

Simulation of FMPC implementation for the car and visual comparison on a desired trajectory:

Animation of a car movement on a different trajectory with obstacles:

Animation of a car movement on a different track without obstacles:

Animation of a car movement on a different track with 1 obstacle:

Animation of a car movement on a different track with 3 obstacles:

---

Run the Project

Step-by-step instructions on how to replicate the results obtained in this project. This should be clear enough for someone with basic knowledge of the tools used to follow.

Requirements

List of prerequisites, dependencies, and environment setup necessary to run the project.

Setup and Installation

Instructions for setting up the project environment, which may include:

• Setting up a virtual environment
• Installing dependencies: pip install -r requirements.txt

Running the Code

Exact commands to execute the project, such as:

python MPC_SIM.py

or alternatively:

python track.py

to launch simulation on the track.

Documentation

• PDF
• Presentation

---

References

• Differential Flatness
• Bicycle Model
• MPC