Our Technologies

Planning is a critical component of any self-driving system. We specialize in developing state-of-the-art technologies for a diverse set of autonomous decision making tasks, including route planning, high-speed motion planning on highways, planning under uncertainty, predicting the intentions of dynamic elements on the road such as other vehicles and pedestrians.

Below are some of our systems in action.

Click to play the high-speed motion planning on the highway video

High-Speed Motion Planning on the Highway

Our planning algorithm generates safe and comfortable motions for self-driving vehicles traveling at high-speeds. Our algorithm can negotiate difficult situations encountered on the highway with high levels of safety, comfortability, and reliability.

Click to play the real-time prediction of intentions video

Real-time Prediction of Intentions

Our prediction algorithms analyze surrounding vehicles, pedestrians, and other dynamic objects in the environment to determine their intentions. Being able to reliably predict future intentions is integral to the performance of a self-driving vehicle.

Click to play the robust planning in uncertain environments video

Robust Planning in Uncertain Environments

Taking into account uncertainty about the intentions of other objects around a self-driving vehicle (e.g. pedestrians, other vehicles, bicyclists) is critical to making safe and comfortable decisions. We have developed novel algorithms that plan efficiently while explicitly reasoning about the uncertainty in the environment surrounding the self-driving vehicle.

Route Planning

Our route planning algorithms generate a long-term plan from a start location to a destination location, taking into account traffic information, road closures, detours, etc. and adjusts the route of the self-driving vehicle accordingly. We employ state-of-the-art search algorithms that allow us to find routes over large road networks in real-time.


Maxim Likhachev

Founder and CEO

Maxim is a Research Assistant Professor with the Robotics Institute and National Robotics Engineering Center (NREC), both part of School of Computer Science, Carnegie Mellon University. He is also an adjunct faculty at the Computer and Information Science department at University of Pennsylvania and a member of the GRASP laboratory.

Ellis Ratner

Lead Robotics Researcher

Ellis develops and implements robust autonomous planning and decision making systems. Prior to joining RobotWits, Ellis was a member of the Search-Based Planning Laboratory (SBPL) at Carnegie Mellon University's Robotics Institute, where he worked on mobile manipulation. Ellis earned a B.A. in Computer Science and Mathematics with honors from Bowdoin College, where he received a national Barry M. Goldwater scholarship.

Lu Lyu

Senior Robotics Researcher

Lu develops prediction algorithms for on-road environments and works on other software engineering tasks in the context of developing software for self-driving vehicles. Prior to joining RobotWits, Lu worked for two start-ups and has hands-on experience developing robotic hardware, perception, and planning systems. Lu received her MS in Robotic Systems Development from Carnegie Mellon University in 2015.

Alla Safonova

Project Manager

Alla brings expertise in computer graphics, interfaces, visualization and robotics. Prior to joining RobotWits, Alla worked as an Assistant Professor at University of Pennsylvania. Alla holds her Ph.D. from Carnegie Mellon University and Master's degree from Georgia Institute of Technology, both in Computer Science.