Heuristic optimization algorithms to improve the performance of a lane detection and tracking algorithm, using Webots simulator.
Find a file
2021-05-12 20:46:25 +02:00
controllers Fine-tuning improvements 2019-06-25 23:58:06 +02:00
plugins/robot_windows First commit 2019-03-19 19:16:08 +01:00
protos Removed unused project resources. 2019-05-23 21:27:08 +02:00
resources/osm_files First commit 2019-03-19 19:16:08 +01:00
results Fine-tuning improvements 2019-06-25 23:58:06 +02:00
worlds Fine-tuning improvements 2019-06-25 23:58:06 +02:00
.gitignore Fine-tuning improvements 2019-06-25 23:58:06 +02:00
importer_demo.sh First commit 2019-03-19 19:16:08 +01:00
LICENSE Initial commit 2019-03-18 20:43:04 +01:00
Makefile First commit 2019-03-19 19:16:08 +01:00
README.md Update README.md 2021-05-12 20:46:25 +02:00

Optimization of a lane detection and tracking algorithm.

This repository contains the implementation of two metaheuristic algorithms developed to optimize an autonomous driving algorithm with the aim of reducing the travel time of a competition circuit.

One of the heuristic algorithms is based on local search engines and taboo lists for optimization. The other heuristics implements a standard genetic algorithm.

Both algorithms are developed in C and run under the Webots simulation environment. Webots provides the simulation engine of the vehicle, the circuit and all the physics.

Installation

  • Webots - Open Source Robot Simulator
  • GSL - GNU Scientific Library
    sudo apt-get install libgsl-dev 
    

Usage

Open and run an instance of the simulator, with the search-based optimization algorithm:

	webots --mode=pause robc_webots/worlds/circuit.wbt

Open and run an instance of the simulator, with the genetic optimization algorithm

	webots --mode=pause robc_webots/worlds/circuit_ga.wbt