Pratik Ingle

Pratik is a robotics researcher working on grasp-free manipulation: moving objects by shaping and controlling soft, modular surfaces instead of gripping them. He is a PhD candidate at the IT University of Copenhagen (REAL Lab) and is currently a visiting researcher at University College London working on tactile sensing and self-supervised learning.

• Papers → Google Scholar or CV
• Talks → See list
• Research → Overview
• Projects → Portfolio
• Blog → Notes and tutorials
• Code → GitHub

Recent: Our preprint on scalable modular manipulation surfaces is on arXiv and was presented at RoboSoft 2026. I spent Mar–Aug 2026 as a visiting PhD at UCL’s CRISP Lab working on tactile sensing. PhD defence expected October 2026.

Publications

Full list on Google Scholar and the research page.

• 2026: Scalable Surface-Based Manipulation Through Modularity and Inter-Module Object Transfer - with J. Lambertsen, K. Støy, A. Faíña. arXiv:2601.21884. Explainer post.
• 2025: Heterogeneous Object Manipulation on Nonlinear Soft Surface Through Linear Controller - IEEE CASE 2025. Explainer post.
• 2025: Soft Manipulation Surface With Reduced Actuator Density For Heterogeneous Object Manipulation - IEEE RoboSoft 2025, oral presentation. Explainer post.

Talks

• RoboSoft 2026 - Scalable Surface-Based Manipulation Through Modularity and Inter-Module Object Transfer.
• CASE 2025 - Heterogeneous object manipulation on nonlinear soft surface through linear controller.
• RoboSoft 2025 - Soft Manipulation Surface With Reduced Actuator Density For Heterogeneous Object Manipulation.
• EI 2025 - Harnessing Fabric Softness and Scalability for Object Manipulation.
• ANTS 2024 - Moving Depot (MOD): An Efficient Depot Motion Strategy for Multi-Robot Foraging.

Research

I work on robotic manipulation surfaces: soft, fabric-based platforms driven by sparse actuators that move heterogeneous and fragile objects without grasping. Current threads:

• Soft manipulation surfaces - reduced actuator density, geometric control, heterogeneous objects.
• Modular manipulation - shared-boundary actuation and inter-module object transfer.
• Tactile sensing - MAE/JEPA representation learning on normal and shear tactile channels, MuJoCo sensor simulation (UCL CRISP Lab).
• Earlier: swarm robotics, multi-robot foraging, and collective decision-making.

Experience

• REAL Lab, IT University of Copenhagen - PhD researcher, Oct 2023–present. Supervisors: Kasper Støy and Andrés Faíña.
• CRISP Lab, University College London - Visiting PhD researcher, Mar–Aug 2026.
• Swarm Lab, NJIT and MOON Lab, IISER Bhopal - collective behaviour modelling, swarm robotics, UAVs (2020–2023).
• IISER Bhopal - Integrated BS-MS, Electrical Engineering and Computer Science, minor in Data Science, 2017–2023.

Selected Projects

See the full portfolio.

• Moving Depot (MOD) - adaptive depot motion for multi-robot foraging (ANTS 2024).
• Decision making in collectives - dynamic individual memory and consensus in changing environments.
• Influencer agents with quantum circuits - leader selection for flock control.
• UVC disinfection UAV - autonomous aerial robot built during COVID-19.

Writing

Recent posts from the blog.

• MAE vs JEPA - two ways to learn representations without labels.
• Normal and shear forces in tactile sensing - why 3-channel tactile sensors matter.
• Simulating a tactile sensor in MuJoCo - notes on sim-based data collection.
• Getting started with Webots - a beginner tutorial.
• VQE from scratch - finding the ground-state energy of a Hamiltonian.

Teaching and Service

• Teaching assistant: Deep Learning, Intelligent Robotics, Computer Vision, Digital Image Processing and Bioimage Analysis (ITU).
• Volunteer: ALIFE 2024, NeurIPS 2021.
• Summer schools: IEEE RAS Multi-Robot Systems (2024), SCIoI/ISAB Embodied Intelligence (2023), COSMOS (2022), PRAIRIE/MIAI AI (2021).

Other

• Contact - prin@itu.dk.
• CV - PDF and overview.
• Google Scholar · GitHub · LinkedIn