[Updates] Self-driving cars, quadrotors, and computing, different faces of motion planning

[Andrew Stone]

I know Cliff and Brandon and a few others of us are on the CS Grad list - but this seemed particularly relevant to Michael's work with "collision-free trajectories":


Begin forwarded message:

> From: colloquia <colloquia at cs.unm.edu>
> Subject: [Colloquia] Reminder: Colloquium (Tomorrow) 4/6
> Date: April 5, 2016 at 2:45:18 PM MDT
> To: csgrad at cs.unm.edu, csundergrad at cs.unm.edu, csfaculty at mail.cs.unm.edu, colloquia at cs.unm.edu
> 
> 
> UNM Computer Science Department Colloquium Series
> Wednesday 4/6
>  
>  
> Self-driving cars, quadrotors, and computing, different faces of motion planning
> 
> Motion planning is the process of finding a collision-free trajectory of a mobile robot, given start and end states. Sensor noise, mechanical constraints, and changing environments make an already challenging problem even more difficult. Lately, artificial intelligence methods, such as reinforcement learning, showed promising results in solving motion planning problems. In this talk, I will present three different motion planning applications. First, an overview of Google’s self-driving car project will highlight motion planning challenges that we must solve when engineering a fully autonomous vehicle. Second, I will talk about quadrotor control with reinforcement learning, as an example of a research project. Finally, I will discuss how we can use motion planning ideas to make computing more adaptive and robust to noise and errors.
>  
> Date:   Wednesday, April 6th
> Time:  2:00-3:00 PM
> Room:  Centennial Engineering Center 1041
> Speaker:  Dr. Aleksandra Faust
> Senior Research Robotics Engineer at X, formerly Google[x]
>  
> This technical talk will be followed by an
> Interview Workshop (3:00-4:00, Centennial Engineering Center 1041) and a
> Social Reception (4:00-5:00, Centennial Engineering Center, Stamm Room)
>  
> Aleksandra Faust is a Senior Research Robotics Engineer at X, formerly Google[x], working on self-driving cars. Prior to joining X, Aleksandra was a Senior Researcher in Sandia National Labs. She earned a Ph.D. in Computer Science (with distinction) at the University of New Mexico, a Masters of Computer Science from University of Illinois at Urbana-Champaign, and a Bachelors in Mathematics from University of Belgrade, Serbia. Aleksandra was the winner of Tom L. Popejoy Award, the best doctoral dissertation at University of New Mexico in Engineering, Mathematics, and Sciences in the period between 2011-2014. She was also awarded with Sandia National Laboratories Doctoral Studies Program and New Mexico Space Grant fellowships, and an Outstanding graduate student in Computer Science award. Her research interests include adaptive planning algorithms for decision making and motion planning, machine learning techniques for solving complex tasks, reinforcement learning, and learning transfer, adaptation and reuse. 
>  
> 
> Faculty Host:  Prof. Lydia Tapia
> Colloquium Faculty contact:  Prof. Tapia, tapia at cs.unm.edu
> Contact the faculty contact indicated above with questions.
> 
> 
> _______________________________________________
> Colloquia mailing list
> Colloquia at mail.cs.unm.edu
> http://mail.cs.unm.edu/cgi-bin/mailman/listinfo/colloquia

Andrew Stone
@twittelator
http://stone.com

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.cocoaconspiracy.com/pipermail/updates/attachments/20160405/30c30edb/attachment.html>