Home research People General Info Seminars Resources Intranet
| Algorithms & Applcations Group | Home | Research | Publications | People | Resources | News
Home Page for Leslie Escalante | Parasol Laboratory

Picture Leslie Escalante
High School/Undergraduate
Algorithms & Applications Group

Parasol Laboratory url: http://parasol.tamu.edu/~lescalante/
Department of Computer Science and Engineering email:
Texas A&M University office: 407 HRBB
College Station, TX 77843-3112 tel:
USA fax: (979) 458-0718

   Howdy, all!

 About Me

I'm Leslie. I currently attend Jimmy Carter Early College High School an am an incoming senior, class of 2015. The "early college" part of our high school offers each student the unique opportunity to graduate twice: with a high school diploma and an Associate degree, earned via the Dual Enrollment program with South Texas College. Although our school aims for an Interdisciplinary degree, I plan for my degree to be in Computer Science.

Meanwhile, this summer, I will work in the Motion Planning project.

 About Our Mentors

Currently, Dr. Nancy M. Amato is a Unocal Professor and Interim Department Head of the Department of Computer Science and Engineering at Texas A&M University. Her research spans motion planning and robotics, computational biology and geometry, and parallel and distributed computing. As of now, Dr. Amato oversees 13 PhD students, 2 masters students, and 10+ undergraduate and high school researchers.

The primary mentor for the undergraduate and high school researchers working to improve Vizmo++ via Haptics and Paths is Read Sandstrom. He is a PhD student who works in the Motion Planning group, under Dr. Amato.


 This summer Juan Aguilar, Ricardo Gonzalez, and I will work with our graduate mentor, Read Sandstrom, in the Vizmo++ group with Path Steering. Since we are three summer research students, we will split the main project into thirds, and each of us will work on making their component run correctly.We will utilize the concept of User-Guided Planning to develop a motion-planning strategy.


 Motion planning, although much improved due to sampling-based planners, is yet to be completely solved: problems with narrow passages, efficiency, and a lack of robustness still remain. User-Guided Planning aims to address these shortcomings by providing a medium for a user to collaborate with the planning algorithm. In this paper, we develop and explore Path Steering, a user-guided technique that lets the user input an approximate path which will serve to seed a sampling-based planner. Path Steering allows the user to bias the sampler to areas of the workspace where a solution is expected to exist, thereby reducing the computation time required to generate efficient samples. Our Path Steering approach can be divided into three major steps: create, where we allow the user to create paths by using a mouse, PHANToM® Haptic device, or the camera; configure, where the planner builds configurations on the user-guided path; connect, where the local planner then attempts to connect valid configurations, and modifies invalid ones. Our approach handles the translation of workspace data into c-space data, so that the path is of use to the planner. Our approach demonstrated an improvement in roadmap and path quality, as well as a decrease in total runtime.

 Read our paper.
 See our poster.