Suneel Belkhale

About

Suneel Belkhale is a robotics researcher whose work spans autonomous aerial systems, robot-assisted feeding, and data-driven learning for robotic manipulation. His early research tackled a fundamental challenge in deep reinforcement learning: how to bridge the sim-to-real gap for vision-based autonomous flight, earning over 114 citations for his work integrating simulated and real-world data. He further advanced aerial robotics by developing model-based meta-reinforcement learning techniques for UAVs transporting suspended payloads, a notoriously difficult control problem, accumulating 94 citations. Belkhale has also made meaningful contributions to robot-assisted feeding, designing systems that balance efficiency and user comfort during bite transfer while incorporating visual and haptic sensing for safe in-mouth food delivery — work with clear humanitarian impact for individuals with mobility impairments. His involvement in DROID, a large-scale in-the-wild robot manipulation dataset (108 citations), reflects his commitment to advancing the data foundations of modern robotics. More recently, he has explored imitation learning data quality and even accelerating diffusion model sampling through parallelization, demonstrating a versatile research profile at the intersection of machine learning and real-world robotic systems.

Research Focus

Key Achievements

Top Papers

1. 1
   Generalization through Simulation: Integrating Simulated and Real Data into Deep Reinforcement Learning for Vision-Based Autonomous Flight

   114 citations · 2019
2. 2
   DROID: A Large-Scale In-The-Wild Robot Manipulation Dataset

   108 citations · 2024
3. 3
   Model-Based Meta-Reinforcement Learning for Flight With Suspended Payloads

   94 citations · 2021
4. 4
   Balancing Efficiency and Comfort in Robot-Assisted Bite Transfer

   16 citations · 2022
5. 5
   In-Mouth Robotic Bite Transfer with Visual and Haptic Sensing

   13 citations · 2023
6. 6
   Generalization through Simulation: Integrating Simulated and Real Data into Deep Reinforcement Learning for Vision-Based Autonomous Flight

   11 citations · 2019
7. 7
   Learning Bimanual Scooping Policies for Food Acquisition

   9 citations · 2022
8. 8
   Parallel Sampling of Diffusion Models

   9 citations · 2023
9. 9
   Learning Visuo-Haptic Skewering Strategies for Robot-Assisted Feeding

   7 citations · 2022
10. 10
    Data Quality in Imitation Learning

    7 citations · 2023

Key Collaborators

Contact & Links