Skip to content
2018-2019 group project in Ecole Polytechnique, robot-taxi fleetsizing
Java
Branch: master
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Type Name Latest commit message Commit time
Failed to load latest commit information.
.settings
bin
demo
src
.classpath
.project
readme.md
result_lambda=1.0MATIN.txt
result_lambda=1.0ORDINARY.txt
result_lambda=1.0SOIR.txt

readme.md

Projet scientifique collectif(PSC)

Sep. 2018 - May. 2019

This project is under the supervision of Renault and Departement of mathematics and Management in École Polytechnique, concerns the optimisation of robot-taxi service(driveless, electrical, sharing, dynamic dispatching). According to the ambition of Renault, we may be able to see robot-taxi serve in Paris in 2022.

Our works focused on the fleet-sizing and pricing, composed of two parts: Mathematical modeling and JAVA simulation. What you see here is the Java simulation.


The model:

The city is segmented into a square grid, clients appear randmoly according to a Poisson Process at each node. A dispatching algorithme is in charge of manoeuvring taxis and deciding how to share a ride between clients.

Some demonstrations of our results:

rate of utilisation

waiting time of client

distribution fluctuation in time

spatial client distribution fluctuation in time

Combining the result of mathematical modeling and this simulation, it turns out that we need 8600 robot-taxis to satisfy all clients instead of the current number 16000!

You can’t perform that action at this time.