Adversarial and Unstructured Environments via Reinforcement Learning

Motion Planning & Decision Making | Perception | Localization | Controls

More About


Make connected autonomous driving technology accessible, affordable and available to everyone! Significantly reduced Annotated data, Safe & Robust, Cost-Effective, and Mapless!

Autonomous Driving

AI Software Platforms and Solutions

Industrial and Warehouse Robotics

We are developing level-5 autonomous driving technology. Our immediate next goal is to perform 100 KM/H autonomous driving demo on Indian roads, showcasing the capabilities of our technology.

Decision making and motion planning accounts for 70% of our R&D, where our research primarily focuses on reinforcement learning, and various areas of theoretical computer science and applied mathematics.

Earlier, we have been proving our technology: (i) in traffic scenarios with stochastic, complex, and often adversarial dynamics; and (ii) in unstructured environmental conditions in India.

Our perception and behaviour planning algorithms make high-definition (high-fidelity) 3D-mapping of environments redundant. We do not require dense maps for automous driving, and can use just GPS maps for end-to-end navigation.

Meet the Founder

Sanjeev Sharma

CEO and Founder

  • Sanjeev initiated his research in autonomous navigation in unknown environments in January 2009, when he was an undergrad student at IIT Roorkee.
  • Since 2014, his research has spanned across several areas of theoretical computer science and applied mathematics, focusing on applications in autonomous driving, including deep learning, computer vision, SLAM, and visual odometry.
  • His research at Swaayatt Robots, to enable autonomous driving in environments as difficult as in India, has been covered by both the national and international media, on several occasions.

Prior Funding and Ecosystem Support