Motion planning for autonomous vehicles is an important topic across the world. Nowadays there exist several classes of algorithms for motion planning, such as grid-based algorithms and sampling-based algorithms. A simple and effective method to plan the path for an autonomous robot based on a YAMAHA Grizzly all terrain vehicle, is here presented, accounting for the anti-rollover problem as well. Grid-based algorithms are a good choice. This thesis presents a few grid-based algorithms and two ways to decide the vehicle commands. A benchmark among different grid-based algorithms and different methods to compute the vehicle commands is shown, together with an analysis of their pros and cons.
Il problema della pianificazione del percorso per veicoli autonomi, ovvero privi di conducente umano è un tema di grande interesse in tutto il mondo. Al giorno d'oggi esistono diverse classi di algoritmi per la pianificazione del moto, quali ad esempio quelli grid-based e sampling-based. In questo lavoro viene presentato un metodo semplice ed efficace per la pianificazione del percorso di un veicolo autonomo basato su un ATV YAMAHA Grizzly. Tale metodo è di tipo grid-based e permette di considerare il problema del ribaltamento del veicolo. Questa tesi illustra alcuni algoritmi grid-based e due modi per decidere l'azione di controllo del veicolo. Sulla base di un problema benchmark verranno analizzati pro e contro di ciascun algoritmo.
Planning dynamic trajectories within the search based planning library
LI, MING
2011/2012
Abstract
Motion planning for autonomous vehicles is an important topic across the world. Nowadays there exist several classes of algorithms for motion planning, such as grid-based algorithms and sampling-based algorithms. A simple and effective method to plan the path for an autonomous robot based on a YAMAHA Grizzly all terrain vehicle, is here presented, accounting for the anti-rollover problem as well. Grid-based algorithms are a good choice. This thesis presents a few grid-based algorithms and two ways to decide the vehicle commands. A benchmark among different grid-based algorithms and different methods to compute the vehicle commands is shown, together with an analysis of their pros and cons.File | Dimensione | Formato | |
---|---|---|---|
2012_12_Li_02.pdf
accessibile in internet per tutti
Descrizione: Manual to compile the library
Dimensione
37.97 kB
Formato
Adobe PDF
|
37.97 kB | Adobe PDF | Visualizza/Apri |
2012_12_Li_03.pdf
accessibile in internet per tutti
Descrizione: Library function manual
Dimensione
51.08 kB
Formato
Adobe PDF
|
51.08 kB | Adobe PDF | Visualizza/Apri |
2012_12_Li_01.pdf
accessibile in internet per tutti
Descrizione: Thesis text
Dimensione
1.17 MB
Formato
Adobe PDF
|
1.17 MB | Adobe PDF | Visualizza/Apri |
I documenti in POLITesi sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
https://hdl.handle.net/10589/72447