πŸš— INGI Dakar 2K21 - Be the first one on the finish line ! πŸš—

Overview

πŸš— INGI Dakar 2K21 - Be the first one on the finish line ! πŸš—

This year's first semester Club Info challenge will put you at the head of a car racing team. You will participate to the world's most famous racing contest, the INGI Dakar. Your goal is to build the best car, and to beat your opponents by reaching the furthest distance from the starting line.

Challenge statement

Each group will be put in charge of a car racing team. Ultimately, your goal is to reach the furthest distance from the starting line, with any of your cars. For this, you will have 6 generations of 20 cars. Each generation will be produced based on the previous one. Your job is thus to implement the algorithm that takes the previous generation of cars in argument, and that produces the next generation. Such an algorithm is called a genetic algorithm, for which a theoretical background is given hereafter.

Genetic algorithms

Genetic algorithms (GA) are inspired by the process of natural selection. They are used to resolve complex problems. They use operators such as mutation, crossover and selection. GA process is split into generations. Each generation is composed of a finite number of individuals which are built from the best individuals of the last generation and one or several operators. The first generation is generally randomly created.

Genetic algorithms are used for a large variety of problems from antenna shape optimization to minimize the weight of structures embarked in mars rovers.

A genetic algorithm is based on three operators:

  • Mutation, a mutation is a random modification of a parameter for an individual in the generation,
  • Crossover, a crossover is the creation of an individual based on parameters values from several members of the last generation,
  • Selection, in a genetic algorithm, we select the best individuals of a generation to construct the next generation.

Alternative text describing the image

The Mutation operator is used to ensure that the selection is not trapped in a local optima and can not reach the global optima for each parameters.

Some useful links:

Program specifications

The program for the INGI Dakar 2K21 is composed of 7 Python modules:

  • Car.py: Defines the class Car that represents a car of the game. A Car is composed of two Wheels and a Chassis, where the Wheels are located on two of the four Chassis vertices.
  • Chassis.py: Defines the class Chassis that represents a car chassis. A Chassis is represented by four vertices connected with each other in a quadrilateral shape.
  • CustomFormatter.py: Used for logging purposes.
  • Game.py: Defines the class Game that represents a game of INGI Dakar 2K21, i.e. the simulation of the 6 generations of 20 cars.
  • main.py: Entry point of INGI Dakar 2K21, which launches the simulations and computes the score.
  • Terrain.py: Defines the class Terrain that represents the terrain on which the cars are driving.
  • Wheel.py: Defines the class Wheel that represents a car's wheel. A Wheel is defined by its radius and the fact that it is a motor wheel or not.

To participate to the challenge, you only have to modify the function next_generation in the module main.py, that takes a representation of the game world (a b2World object) and the previous generation of cars (a list of Car objects) as arguments, and that returns the next generation of cars (also a list of Car objects). The car features that you can update for the next generation are given below.

Car features

A car is composed of the following (the numbers in bold cannot be changed):

  • TWO wheels, one of which is a motor wheel
  • A chassis, composed by FOUR vertices, linked together to form a polygon shape.

The car features that you can modify to reach the maximum distance are the following:

  • Radius of the two wheels, separately.
  • Which wheel is the motor wheel.
  • Position of the four vertices of the chassis.
  • To which of the chassis' vertices the two wheels are attached.

Please consult the corresponding classes to understand how those features are expressed and used in the program.

Score computation

To start the simulation of the challenge, just run the python3 main.py Python module. You must also activate the python virtual environment with source penv/bin/activate.

The execution of the challenge, and computation of your final score, is as follows:

  • Each generation contains 20 cars. The maximum distance reached by any of the cars is recorded as the score of this generation.
  • A game is composed of 6 generations. The score of a game is the maximum score among all the generations.
  • To smoothen the results, 5 games are launched after each other. Your final score is the average of the different score you obtained during the games.

At the end of the 5 games, a plot will be shown, with your results for the 5 games.

Installation and execution

Installation

To install the project, first clone the repository with the following command:

git clone https://github.com/ClubINFO-INGI-UCLouvain/INGI-Dakar-2K21-Challenge.git

Then, install the needed libraries by running the install.sh script, inside the project directory:

python3 -m venv penv;
source  penv/bin/activate;
chmod +x install.sh;
./install.sh;

Execution

To run the challenge simulation, you can simply run the main.py Python module in the src directory, with the following command:

cd src/
python3 main.py [--seed_terrain SEED] [--seed_car SEED] [--no_UI] [--no_plot]

The command line arguments, all optional, are the following:

  • --seed_terrain SEED (with SEED an integer): sets the seed for the random generation of the game terrain to SEED, for reproducibility of the simulations
  • --seed_car SEED (with SEED an integer): sets the seed for the random generation of the first generation of cars to SEED, for reproducibility of the simulations
  • --no_UI: does not show the graphical interface of the game, which drastically speeds up the simulations
  • --no_plot: does not show the plot of the games' result at the end of all the games

Note that, for the contest, the seeds will be fixed for equity among the groups.

There is also a hidden argument, maybe you can try to find it πŸ˜‰

Owner
ClubINFO INGI (UCLouvain)
ClubINFO INGI (UCLouvain)
Python implementation of Bayesian optimization over permutation spaces.

Bayesian Optimization over Permutation Spaces This repository contains the source code and the resources related to the paper "Bayesian Optimization o

Aryan Deshwal 9 Dec 23, 2022
Official repository for Few-shot Image Generation via Cross-domain Correspondence (CVPR '21)

Few-shot Image Generation via Cross-domain Correspondence Utkarsh Ojha, Yijun Li, Jingwan Lu, Alexei A. Efros, Yong Jae Lee, Eli Shechtman, Richard Zh

Utkarsh Ojha 251 Dec 11, 2022
ViSER: Video-Specific Surface Embeddings for Articulated 3D Shape Reconstruction

ViSER: Video-Specific Surface Embeddings for Articulated 3D Shape Reconstruction. NeurIPS 2021.

Gengshan Yang 59 Nov 25, 2022
StackGAN: Text to Photo-realistic Image Synthesis with Stacked Generative Adversarial Networks

StackGAN Pytorch implementation Inception score evaluation StackGAN-v2-pytorch Tensorflow implementation for reproducing main results in the paper Sta

Han Zhang 1.8k Dec 21, 2022
TLDR: Twin Learning for Dimensionality Reduction

TLDR (Twin Learning for Dimensionality Reduction) is an unsupervised dimensionality reduction method that combines neighborhood embedding learning with the simplicity and effectiveness of recent self

NAVER 105 Dec 28, 2022
An implementation of Geoffrey Hinton's paper "How to represent part-whole hierarchies in a neural network" in Pytorch.

GLOM An implementation of Geoffrey Hinton's paper "How to represent part-whole hierarchies in a neural network" for MNIST Dataset. To understand this

50 Oct 19, 2022
Multi-Glimpse Network With Python

Multi-Glimpse Network Multi-Glimpse Network: A Robust and Efficient Classification Architecture based on Recurrent Downsampled Attention arXiv Require

9 May 10, 2022
A Shading-Guided Generative Implicit Model for Shape-Accurate 3D-Aware Image Synthesis

A Shading-Guided Generative Implicit Model for Shape-Accurate 3D-Aware Image Synthesis Project Page | Paper A Shading-Guided Generative Implicit Model

Xingang Pan 115 Dec 18, 2022
Animatable Neural Radiance Fields for Modeling Dynamic Human Bodies

To make the comparison with Animatable NeRF easier on the Human3.6M dataset, we save the quantitative results at here, which also contains the results of other methods, including Neural Body, D-NeRF,

ZJU3DV 359 Jan 08, 2023
Source code of the paper Meta-learning with an Adaptive Task Scheduler.

ATS About Source code of the paper Meta-learning with an Adaptive Task Scheduler. If you find this repository useful in your research, please cite the

Huaxiu Yao 16 Dec 26, 2022
Official PyTorch implementation of "Evolving Search Space for Neural Architecture Search"

Evolving Search Space for Neural Architecture Search Usage Install all required dependencies in requirements.txt and replace all ..path/..to in the co

Yuanzheng Ci 10 Oct 24, 2022
PyTorch-based framework for Deep Hedging

PFHedge: Deep Hedging in PyTorch PFHedge is a PyTorch-based framework for Deep Hedging. PFHedge Documentation Neural Network Architecture for Efficien

139 Dec 30, 2022
Cognition-aware Cognate Detection

Cognition-aware Cognate Detection The repository which contains our code for our EACL 2021 paper titled, "Cognition-aware Cognate Detection". This wor

Prashant K. Sharma 1 Feb 01, 2022
Code for the paper "Controllable Video Captioning with an Exemplar Sentence"

SMCG Code for the paper "Controllable Video Captioning with an Exemplar Sentence" Introduction We investigate a novel and challenging task, namely con

10 Dec 04, 2022
No-reference Image Quality Assessment(NIQA) Algorithms (BRISQUE, NIQE, PIQE, RankIQA, MetaIQA)

No-Reference Image Quality Assessment Algorithms No-reference Image Quality Assessment(NIQA) is a task of evaluating an image without a reference imag

Dae-Young Song 26 Jan 04, 2023
πŸ”₯πŸ”₯High-Performance Face Recognition Library on PaddlePaddle & PyTorchπŸ”₯πŸ”₯

face.evoLVe: High-Performance Face Recognition Library based on PaddlePaddle & PyTorch Evolve to be more comprehensive, effective and efficient for fa

Zhao Jian 3.1k Jan 04, 2023
The dataset of tweets pulling from Twitters with keyword: Hydroxychloroquine, location: US, Time: 2020

HCQ_Tweet_Dataset: FREE to Download. Keywords: HCQ, hydroxychloroquine, tweet, twitter, COVID-19 This dataset is associated with the paper "Understand

2 Mar 16, 2022
Learning Efficient Online 3D Bin Packing on Packing Configuration Trees

Learning Efficient Online 3D Bin Packing on Packing Configuration Trees This repository is being continuously updated, please stay tuned! Any code con

86 Dec 28, 2022
Course about deep learning for computer vision and graphics co-developed by YSDA and Skoltech.

Deep Vision and Graphics This repo supplements course "Deep Vision and Graphics" taught at YSDA @fall'21. The course is the successor of "Deep Learnin

Yandex School of Data Analysis 160 Jan 02, 2023