PORTALE DELLA DIDATTICA

Ricerca CERCA
  KEYWORD

DAUIN - GR-06 - ELECTRONIC DESIGN AUTOMATION - EDA

Optimization of the path of a vehicle for the transportation of materials in an industrial plant using artificial intelligence algorithms

azienda Thesis in external company    


keywords ARTIFICIAL INTELLIGENCE, INDUSTRIAL AUTOMATION, INDUSTRY 4.0, LOGISTICS, OPTIMIZATION

Reference persons DANIELE JAHIER PAGLIARI

External reference persons Ing. Michele D'Onghia (RADA Soluzioni Informatiche)

Research Groups DAUIN - GR-06 - ELECTRONIC DESIGN AUTOMATION - EDA, ELECTRONIC DESIGN AUTOMATION - EDA, GR-06 - ELECTRONIC DESIGN AUTOMATION - EDA

Thesis type EXPERIMENTAL, SOFTWARE DEVELOPMENT, STAGE + THESIS

Description Automated Guided Vehicles (AGV) are robots used in industrial plants for the movement of different kinds of materials and products. The optimization of the path followed by these vehicles, aimed at delivering materials to different stations of a production line in the minimum amount of time, has important implications on the entire plant logistics.

The goal of this thesis is to implement such optimization in the real scenario offered by an important multinational company in the automotive sector. The candidate will therefore initially build a formal model of the problem, taking into account all constraints imposed by the plant (obstacles, speed limits in certain areas, “one-way paths”, etc). Then, the candidate will deal with the optimization itself by developing appropriate artificial intelligence algorithms. Specifically, the candidate will have the opportunity of analyzing and comparing both classical algorithms based on graph theory, and data-driven solutions. Lastly, the candidate will have the opportunity of validating his solution both using a CAD simulation and on the field.

The thesis project will be mainly carried out at the headquarters of RADA, a software solutions company, in Rivoli (TO). There will be a reimbursement of expenses.

Required skills Sono richieste buone capacità di programmazione e problem solving. È inoltre utile una buona conoscenza dei problemi di ottimizzazione e degli algoritmi di ricerca basati su grafi. Infine, può essere d’aiuto, anche se non è da considerarsi necessaria, una minima familiarità con concetti di machine/deep learning.


Deadline 30/08/2021      PROPONI LA TUA CANDIDATURA




© Politecnico di Torino
Corso Duca degli Abruzzi, 24 - 10129 Torino, ITALY
Contatti