activated
Home
build details

# Task: Autonomous Mobility-on-Demand on AMoDeus

Modified 2018-06-17 by Andrea Censi

TODO for Claudio Ruch: Please match notation to overall document - cost function should be $\objective$

previous task next (9 of 24) index

The following was marked as "todo".

TODO for Claudio Ruch: Please match notation to overall document - cost function should be $\objective$

in repo duckietown/docs-AIDO branch master commit 4e676b04

Created by function create_notes_from_elements in module mcdp_docs.task_markers.

In this task, we zoom out of Duckietown and switch to a task so big that it is not yet accessible in Duckietown but only in simulation. This is likely changing as Duckietowns across the world experience a tremendous growth in population, partly due to the surprising fertility rate of the Duckies themselves, partly due to the significant immigration of Duckies of engineering background.

In this task, we will therefore use the simulation platform AMoDEUs [19] to orchestrate thousands of robotic taxis in cities to pick up and deliver customers in this virtual autonomous mobility-on-demand system composed of robotaxis.

The design of operational policies for AMoD systems is a challenging tasks for various reasons that make it also very suitable as a machine learning problem:

• The vehicles run on a large and complex network with varying travel-times and time-varying congestion effects.
• Every decision taken influences other decisions, e.g., if a specific car is sent to pick up some request, it is not available anymore to service another request close-by.
• To run an AMoD system with good service levels efficiently is a task that requires to balance the number of employed robotic taxis, the amount of rebalancing being conducted and the statistical distribution of waiting times.

Operating AMoD systems is a challenging task in which many conventional algorithms have failed, this makes it ideal as a challenge for NIPS!

No questions found. You can ask a question on the website.