I will be joining Georgia Tech's George W. Woodruff School of Mechanical Engineering as an Assistant Professor, starting Jan. 2023. I'm going home!

I'm Shreyas, and I study how to make autonomous systems, such as robots, safe and practical.

I'm currently a postdoctoral scholar at Stanford University, working in the ASL under Prof. Marco Pavone. I completed my Ph.D. in Mechanical Engineering at the University of Michigan, advised by Prof. Ram Vasudevan in the ROAHM Lab. I studied Mechanical Engineering in undergrad at Georgia Tech, and was advised by Prof. Antonia Antoniou.

Contact: skousik(at)stanford.edu

Latest News

16 June 2022: It's been a great two weeks, with three papers accepted! TAC, RA-L, and ITSC.

1 June 2022: I accepted my offer to be an assistant professor at Georgia Tech!

3 Mar 2022: I'm serving as an associate editor for IROS 2022.

3 Jan 2022: I'm helping organize the ICRA Robotics Debates this year. Come check it out!

21 Dec 2021: Our full-day workshop, "Motion Planning with Implicit Neural Representations of Geometry," was accepted to ICRA 2022!


See all past updates here.

About Me

I work on bridging the gap between theory and application to ensure that formal safety guarantees are practical to implement on real-world systems. I use techniques from control theory, robot motion planning, and data-driven design.

I’m from Charleston, South Carolina. I received a B.S. in Mechanical Engineering from Georgia Tech in December 2014, and M.S. and Ph.D. degrees in Mechanical Engineering from the University of Michigan in 2020. I'm currently a postdoctoral scholar at Stanford. For more details, check out my CV.

I also read a lot of sci-fi, with a focus on authors from underrepresented populations in sci-fi and STEM. You can read about my literary adventures here, and check out some slides about the history and role of sci-fi here.

Research

My research is focused on guaranteeing safety in autonomy, which I usually take to mean collision avoidance for robots. The key challenge I address is to translate safety in math to safety on real robots.

To this end, I created Reachability-based Trajectory Design, or RTD. This is a method for autonomously generating dynamically-feasible, collision-free trajectories for a variety of mobile robot morphologies. Check out the tutorial for a walkthrough.

Currently, I am exploring ways to model uncertainty from autonomous perception and estimation systems, and ensure that these models are practical for downstream planning and control tasks.

To read more about my work, and check out videos, click here.

Teaching and Outreach

One of the main reasons I have stayed in academia is to teach and mentor students. I first realized my passion for teaching with Georgia Tech's Public Speaking Club. As a volunteer and graduate student instructor, I've continued pursuing opportunities to spread STEM in Michigan and at Stanford.

While at Michigan, I worked with the Michigan Engineering Zone, Michigan Xplore Engineering, and the REACT workshop. I have also helped teach the ROB 599 Autonomous Vehicles course, and have mentored several undergrad and Master's students.

Here at Stanford, my labmate Derek Knowles gave me the chance to teach kids (over Zoom) about soft robots and how math is music for the Stanford Science Bus. I've also recently volunteered with AI4ALL.

Diversity, Equity, and Inclusion

In the long term, I see autonomous systems as a way to increase equity in society, because they have the potential to help people across boundaries of ability or advantage. As I have found in my outreach work, robots in particular are a great way to bring a wide variety of people together in inclusive environments.

But, it's also critical that we as engineers do the hard work now of creating diverse and inclusive communities that strive towards equitable applications of technology. To this end, I'm currently working on a few community-building efforts within Stanford and through robotics. I am eager to discuss these topics and efforts, so feel free to send me an email!

Selected Publications

  1. S. Kousik*, S. Vaskov*, F. Bu, M. Johnson-Roberson, and R. Vasudevan. “Bridging the gap between safety and real-time performance in receding-horizon trajectory design for mobile robots.” The International Journal of Robotics Research, September 2020. [link].

  2. S. Kousik, P. Holmes,and R. Vasudevan. “Safe, aggressive quadrotor flight via reachability-based trajectory design.” DSCC 2019. [link].

  3. P. Holmes, S. Kousik, B. Zhang, D. Raz, C. Barbalata, M. Johnson-Roberson, and R. Vasudevan. "Reachable sets for safe, real-time manipulator trajectory design." R:SS 2020. [link].