Kavan Singh Sikand

About

Kavan Singh Sikand is a leading researcher in autonomous mobile robotics, with a focus on high-speed off-road navigation, multi-robot fleet management, and preference-aware path planning. His most impactful work, "VI-IKD" (31 citations), introduces a learned visual-inertial inverse kinodynamics model that enables ground vehicles to navigate challenging off-road terrain at high speeds by accounting for complex vehicle-terrain interactions. Sikand further advanced robot control with a method combining learned forward kinodynamics and non-linear least squares optimization (17 citations), achieving accurate high-speed control. In fleet robotics, he developed Robofleet (24 citations), an open-source communication and management framework that ensures reliable, secure robot-human interaction during long-term deployments. His research on visual representation learning for preference-aware path planning (17 citations) allows robots to reason about terrain safety and operator preferences, such as avoiding mud or flower beds. Sikand also contributed to last-mile delivery with an open-source framework for heterogeneous robot teams. His work bridges learning-based kinodynamics and practical deployment, earning recognition for enabling robust, real-world autonomous systems.

Research Focus

Key Achievements

Top Papers

1. 1
   VI-IKD: High-Speed Accurate Off-Road Navigation using Learned Visual-Inertial Inverse Kinodynamics

   31 citations · 2022
2. 2
   Robofleet: Open Source Communication and Management for Fleets of Autonomous Robots

   24 citations · 2021
3. 3
   Visual Representation Learning for Preference-Aware Path Planning

   17 citations · 2022
4. 4
   High-Speed Accurate Robot Control using Learned Forward Kinodynamics and Non-linear Least Squares Optimization

   17 citations · 2022
5. 5
   An Open-Source Framework for Last Mile Delivery with Heterogeneous Robots

   3 citations · 2021

Key Collaborators

Contact & Links