IPAM Workshop on Autonomous Vehicles

Details about the world of Applied Mathemathics

Image credit: Borok Times

Last week I had the amazing opportunity to visit UCLA. My visit was in the framework of a workshop in Autonomous Vehicles offered by IPAM (Institute of Pure and Applied Mathemathics). In 2015, I had the opportunity to attend to the long program on New Directions in Mathematical Approaches for Traffic Flow Management. I’ve always thought Los Angeles is the perfect city to speak about traffic given the volume of vehicles that can be found. Moreover congestion is a big problem in this city according to recent reports here. At that time the focus was really on Traffic Flow Theory and how technology could help to improve traffic conditions. A summary of speakers and the lectures offered at that time can be found here.

This time there were interesting solutions to the congestion problem, moreover, the approach was really different, in fact the future of transportation seems to be automated more and more and Autonomous Vehicles are being part of this alternative revolution to solve most of the congestion problems. Although this might be the case, a lot of questions are still unsolved, some will emerge. For example, will we use more the vehicle if it is an autonomous one? As Dan Work pointed out: “From the moment we have Netflix in our cars our transportation habits will probably change, we’ll drive more, meaning more vehicles miles travelled” the actual status of development is an undergoing process and still there is a lot to see in the future.

I will list my favorite talks down here so if you have some time, I’d really suggest to give them a look.

  • Checkout the amazing work from Dan and collaborators at evaluation of Adaptive Cruise Control in regular vehicles.here

  • Checkout the flow framework in Reinforcement Learning and its application to traffic scenarios by Cathy here

  • Given that I like like the control approach to these solutions I’ve really appreciated the talk from Ramanarayan here where he speaks about how to study safety in autonomous vehicles from the control point of view, which is of big concern.

Other talks I’ve really appreciated came from the industry side so thanks to Luc (Lyft) and Raquel (Uber) for providing highlights on the directions followed by these companies in terms of autnomous vehicles. I apologize, these talks are not available online. You may find also other wonderful speakers that presented their work. Please find their presentations here.

Having nothing more to say, the journey is still long, I’m looking forward to the upcoming future where autonomous vehicles may have more and more impact in our lives.

Andrés Ladino
Andrés Ladino
Data Scientist

My research interests include distributed control, networked control systems and machine learning.