Ram Vasudevan

About

Ram Vasudevan is a prominent robotics and control systems researcher whose work spans autonomous systems, soft robotics, and data-driven control. His research has made significant strides in bridging theoretical foundations with practical robotic applications, earning him considerable recognition across multiple domains. Vasudevan's most influential contribution applies Koopman operator theory to soft robot control, garnering 283 citations by offering a mathematically rigorous framework for managing the inherently complex dynamics of soft robotic systems. Complementing this, his continuum modeling of fiber-reinforced soft actuators further advances the field's mechanical foundations. In autonomous driving and navigation, his work on pedestrian trajectory prediction using bi-directional variational autoencoders (184 citations) and LiDAR-camera sensor fusion demonstrates a sophisticated approach to safe robot-environment interaction. His multi-robot SLAM paper (195 citations) provides robust solutions for collaborative mapping in GPS-denied environments. Earlier foundational work on convex optimization for nonlinear feedback controllers (143 citations) established theoretical tools now widely adopted for underactuated robotic systems, while his human-inspired bipedal walking cost functions reflect a commitment to biomechanically meaningful robotics. His research into synthetic data generation for deep learning training further highlights his forward-thinking approach to data efficiency in autonomous systems. Across these contributions, Vasudevan has established himself as a versatile and impactful force in modern robotics research.

Research Focus

Key Achievements

Top Papers

1. 1
   Data-Driven Control of Soft Robots Using Koopman Operator Theory

   283 citations · 2020
2. 2
   Pairwise Consistent Measurement Set Maximization for Robust Multi-Robot Map Merging

   195 citations · 2018
3. 3
   BiTraP: Bi-Directional Pedestrian Trajectory Prediction With Multi-Modal Goal Estimation

   184 citations · 2021
4. 4
   Convex optimization of nonlinear feedback controllers via occupation measures

   143 citations · 2014
5. 5
   Driving in the Matrix: Can virtual worlds replace human-generated annotations for real world tasks?

   88 citations · 2017
6. 6
   Human-data based cost of bipedal robotic walking

   49 citations · 2011
7. 7
   A Continuum Model for Fiber-Reinforced Soft Robot Actuators

   45 citations · 2018
8. 8
   LiStereo: Generate Dense Depth Maps from LIDAR and Stereo Imagery

   38 citations · 2020
9. 9
   These Maps are Made for Walking: Real-Time Terrain Property Estimation for Mobile Robots

   30 citations · 2022
10. 10
    Pedestrian Planar LiDAR Pose (PPLP) Network for Oriented Pedestrian Detection Based on Planar LiDAR and Monocular Images

    29 citations · 2019

Key Collaborators

Contact & Links