Portrait of Gregory Dudek is unavailable

Gregory Dudek

Associate Academic Member
Full Professor and Research Director of Mobile Robotics Lab, McGill University
Vice President and Lab Head of AI Research, Samsung AI Center in Montréal

Biography

Gregory Dudek is a full professor at McGill University’s CIM which is linked to the School of Computer Science and Research Director of Mobile Robotics Lab. He is also the Lab Director and VP of research at Samsung AI Center Montreal and an Associate academic member at Mila - Quebec Institute of Artificial Intelligence.

Dudek has authored and co-authored over 300 research publications on a wide range of subjects, including visual object description, recognition, RF localization, robotic navigation and mapping, distributed system design, 5G telecommunications and biological perception.

He co-authored the book “Computational Principles of Mobile Robotics” (Cambridge University Press) with Michael Jenkin. He has chaired and been involved in numerous national and international conferences and professional activities concerned with robotics, machine sensing and computer vision.

Dudek’s research interests include perception for mobile robotics, navigation and position estimation, environment and shape modelling, computational vision and collaborative filtering.

Current Students

PhD - McGill University
Principal supervisor :
Master's Research - McGill University
Principal supervisor :

Publications

Contractive Diffusion Policies: Robust Action Diffusion via Contractive Score-Based Sampling with Differential Equations
Charlotte Morissette
Anas El Houssaini
Diffusion policies have emerged as powerful generative models for offline policy learning, whose sampling process can be rigorously characte… (see more)rized by a score function guiding a Stochastic Differential Equation (SDE). However, the same score-based SDE modeling that grants diffusion policies the flexibility to learn diverse behavior also incurs solver and score-matching errors, large data requirements, and inconsistencies in action generation. While less critical in image generation, these inaccuracies compound and lead to failure in continuous control settings. We introduce Contractive Diffusion Policies (CDPs) to induce contractive behavior in the diffusion sampling dynamics. Contraction pulls nearby flows closer to enhance robustness against solver and score-matching errors while reducing unwanted action variance. We develop an in-depth theoretical analysis along with a practical implementation recipe to incorporate CDPs into existing diffusion policy architectures with minimal modification and computational cost. We evaluate CDPs for offline learning by conducting extensive experiments in simulation and real-world settings. Across benchmarks, CDPs often outperform baseline policies, with pronounced benefits under data scarcity.
MANSION: Multi-floor lANguage-to-3D Scene generatIOn for loNg-horizon tasks
Lirong Che
Shuo Wen
Shan Huang
Chuang Wang
Yuzhe Yang
Xueqian Wang
Jian Su
Real-world robotic tasks are long-horizon and often span multiple floors, requiring complex spatial reasoning. Existing embodied benchmarks,… (see more) however, are largely confined to single-floor homes, failing to evaluate agents on realistic, building-scale tasks. We introduce MANSION, a language-driven framework for generating building-scale, multi-floor 3D environments for long-horizon tasks. Using this framework, we release MansionWorld, a large-scale dataset featuring over 1,000 diverse, non-residential buildings. These environments support cross-floor skills and long-horizon task generation on reusable building layouts. Experiments show that current methods degrade sharply on our multi-floor tasks, highlighting both the challenge and the value of this setting for advancing embodied AI.
On the Coverage of Planar Worlds with Simulated Mobile Ad Hoc Networks
Shuo Wen
Louis-Roy Langevin
Antonio Lorı́a
In this paper, we study the online deployment of mobile ad hoc networks (MANETs) to achieve complete sensing coverage in initially unknown p… (see more)lanar worlds while maintaining connectivity, as formalized by the Partially Observable Cooperative Guard Art Gallery Problem (POCGAGP). While prior work provided strong guarantees in orthogonal worlds, many realistic worlds exhibit arbitrary polygonal or curvilinear boundaries that violate these assumptions. We extend the POCGAGP framework and evaluation to such settings and show that the target-selection step of the DADENCE algorithm can be implemented as a neighbor-only average-consensus process. Through evaluation on over 1,500 test cases spanning a wide range of geometric complexities, we demonstrate that DADENCE achieves superior robustness and agent efficiency compared to five benchmark algorithms across both polygonal and curvilinear worlds.
The Surprising Difficulty of Search in Model-Based Reinforcement Learning
Wei-Di Chang
Mikael Henaff
Brandon Amos
This paper investigates search in model-based reinforcement learning (RL). Conventional wisdom holds that long-term predictions and compound… (see more)ing errors are the primary obstacles for model-based RL. We challenge this view, showing that search is not a plug-and-play replacement for a learned policy. Surprisingly, we find that search can harm performance even when the model is highly accurate. Instead, we show that mitigating distribution shift matters more than improving model or value function accuracy. Building on this insight, we identify key techniques for enabling effective search, achieving state-of-the-art performance across multiple popular benchmark domains.
On Mobile Ad Hoc Networks for Coverage of Partially Observable Worlds
Shuo Wen
Louis-Roy Langevin
Antonio Lor'ia
Learning Heuristics for Transit Network Design and Improvement with Deep Reinforcement Learning
Andrew Holliday
Ahmed El-Geneidy
Large Pre-Trained Models for Bimanual Manipulation in 3D
Generalizable Imitation Learning Through Pre-Trained Representations
Wei-Di Chang
Francois Hogan
In this paper we leverage self-supervised vision transformer models and their emergent semantic abilities to improve the generalization abil… (see more)ities of imitation learning policies. We introduce BC-ViT, an imitation learning algorithm that leverages rich DINO pre-trained Visual Transformer (ViT) patch-level embeddings to obtain better generalization when learning through demonstrations. Our learner sees the world by clustering appearance features into semantic concepts, forming stable keypoints that generalize across a wide range of appearance variations and object types. We show that this representation enables generalized behaviour by evaluating imitation learning across a diverse dataset of object manipulation tasks. Our method, data and evaluation approach are made available to facilitate further study of generalization in Imitation Learners.
Learning active tactile perception through belief-space control
Johanna Hansen
Francois Hogan
Robot operating in an open world can encounter novel objects with unknown physical properties, such as mass, friction, or size. It is desira… (see more)ble to be able to sense those property through contact-rich interaction, before performing downstream tasks with the objects. We propose a method for autonomously learning active tactile perception policies, by learning a generative world model leveraging a differentiable bayesian filtering algorithm, and designing an information- gathering model predictive controller. We test the method on three simulated tasks: mass estimation, height estimation and toppling height estimation. Our method is able to discover policies which gather information about the desired property in an intuitive manner.
AIoT Smart Home via Autonomous LLM Agents
Dmitriy Rivkin
Francois Hogan
Amal Feriani
Adam Sigal
Xue Liu
A Blockchain Framework for Equitable and Secure Task Allocation in Robot Swarms
Alexandre Pacheco
Xue Liu
Marco Dorigo
Recent studies demonstrate the potential of blockchain to enable robots in a swarm to achieve secure consensus about the environment, partic… (see more)ularly when robots are homogeneous and perform identical tasks. Typically, robots receive rewards for their contributions to consensus achievement, but no studies have yet targeted heterogeneous swarms, in which the robots have distinct physical capabilities suited to different tasks. We present a novel framework that leverages domain knowledge to decompose the swarm mission into a hierarchy of tasks within smart contracts. This allows the robots to reach a consensus about both the environment and the action plan, allocating tasks among robots with diverse capabilities to improve their performance while maintaining security against faults and malicious behaviors. We refer to this concept as equitable and secure task allocation. Validated in Simultaneous Localization and Mapping missions, our approach not only achieves equitable task allocation among robots with varying capabilities, improving mapping accuracy and efficiency, but also shows resilience against malicious attacks.
Multimodal and Force-Matched Imitation Learning with a See-Through Visuotactile Sensor
Trevor Ablett
Oliver Limoyo
Adam Sigal
Jonathan Kelly
Francois Hogan
Kinesthetic Teaching is a popular approach to collecting expert robotic demonstrations of contact-rich tasks for imitation learning (IL), bu… (see more)t it typically only measures motion, ignoring the force placed on the environment by the robot. Furthermore, contact-rich tasks require accurate sensing of both reaching and touching, which can be difficult to provide with conventional sensing modalities. We address these challenges with a See-Through-your-Skin (STS) visuotactile sensor, using the sensor both (i) as a measurement tool to improve kinesthetic teaching, and (ii) as a policy input in contact-rich door manipulation tasks. An STS sensor can be switched between visual and tactile modes by leveraging a semi-transparent surface and controllable lighting, allowing for both pre-contact visual sensing and during-contact tactile sensing with a single sensor. First, we propose tactile force matching, a methodology that enables a robot to match forces read during kinesthetic teaching using tactile signals. Second, we develop a policy that controls STS mode switching, allowing a policy to learn the appropriate moment to switch an STS from its visual to its tactile mode. Finally, we study multiple observation configurations to compare and contrast the value of visual and tactile data from an STS with visual data from a wrist-mounted eye-in-hand camera. With over 3,000 test episodes from real-world manipulation experiments, we find that the inclusion of force matching raises average policy success rates by 62.5%, STS mode switching by 30.3%, and STS data as a policy input by 42.5%. Our results highlight the utility of see-through tactile sensing for IL, both for data collection to allow force matching, and for policy execution to allow accurate task feedback.