Skip to content

poludmik/Path-planning-for-multiple-UAVs

Repository files navigation

An Improved RRT* Algorithm for Multi-Robot Path Planning, Bachelor Thesis Project

This repository contains my bachelor thesis project, which I was writing at FEE CTU (FEL ČVUT) in Prague. The name of my bachelor program is Cybernetics and Robotics 🦾. This README provides only a meager view on my whole work, that is full of details and nuances. That said, I encourage you to take a look at my thesis text, located as a LaTeX zip in a directory or on Semantic Scholar as a PDF doc.

Abstract

This thesis includes a brief overview of the UAV path planning and a detailed explanation of the algorithms implemented in C++. The implementation of the RRT and RRT* algorithms were carried out and extended to handle the generation of trajectories for multiple drones. Two obstacle avoidance approaches were introduced and tested with both RRT family path-planning algorithms. Experiments of autonomous UAV flight in a forest-like environment were conducted in both simulation and real life; for this purpose, detection and mapping of trees using an active lidar sensor was implemented.

Path planning algorithms

Repository contains two basic algorithms for path planning in 3D: RRT and RRT*. Parameters of both algorithms were thoroughly explored and compared. Examples of path finding in 2D:

These algorithms were expanded to consider trajectory planning for multiple agents, 2D example:

Obstacle avoidance

To perform obstacle avoidance, the collision detection module was written. For simplicity, two object forms were introduced: spheres and cylinders. However, with small additions, the above path planning would work with more complex objects, i.e. 3D polygon meshes.

Autonomous flight 🛸

It was decided to conduct an autonomous flight experiment to proof the usability and robustness of implemented algorithms. Firstly in simulations and then with a real drone. For a drone to fly through an environment with obstacles, the obstacle detection module is needed. I used a 3D lidar sensor to detect and map the environment around the drone. Videos of both simulation and real life tests are available on my YouTube channel.

A few words about licensing. This is a CTU thesis, so here are the rules CTU provides:
"A university thesis is a work protected by the Copyright Act. Extracts, copies and transcripts of the thesis are allowed for personal use only and at one?s own expense. The use of thesis should be in compliance with the Copyright Act http://www.mkcr.cz/assets/autorske-pravo/01-3982006.pdf and the citation ethics http://knihovny.cvut.cz/vychova/vskp.html"

About

Bachelor thesis project on autonomous path-planning for drones. Path planning is based on RRT family algorithms, and the motion/detection modules are connected with a CTU MRS system.

Topics

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published