11 papers
Advantage-Driven Explicit Memory for Social Navigation
Yeonsoo Park, Mattia Racca, Guillaume Bono +4
Robot policies are predominantly learned with classical parametric variants of imitation learning or RL, where training stores the agent's behavior exclusively in the policy's netw…
Compressing History into Memory: Distilling Transformers into Recurrent Transformers
Philippe Weinzaepfel, Christian Wolf, Bülent Mert Sariyildiz +3
Transformers are AI's workhorse but their computational cost becomes prohibitive when processing long sequences. We target long-horizon streaming vision and robotics applications,…
A scalar per patch from pre-trained ViTs enables fast moving navigation in the real world
Steeven Janny, Leonid Antsfeld, Christian Wolf
Trained policies for real-world robotics rely on computer vision components, typically in the form of pre-trained visual encoders. These encoders are an essential component and it…
Kinaema: a recurrent sequence model for memory and pose in motion
Mert Bulent Sariyildiz, Philippe Weinzaepfel, Guillaume Bono +2
One key aspect of spatially aware robots is the ability to "find their bearings", ie. to correctly situate themselves in previously seen spaces. In this work, we focus on this part…
What does really matter in image goal navigation?
Gianluca Monaci, Philippe Weinzaepfel, Christian Wolf
Image goal navigation requires two different skills: firstly, core navigation skills, including the detection of free space and obstacles, and taking decisions based on an internal…
Reasoning in visual navigation of end-to-end trained agents: a dynamical systems approach
Steeven Janny, Hervé Poirier, Leonid Antsfeld +6
Progress in Embodied AI has made it possible for end-to-end-trained agents to navigate in photo-realistic environments with high-level reasoning and zero-shot or language-condition…
RANa: Retrieval-Augmented Navigation
Gianluca Monaci, Rafael S. Rezende, Romain Deffayet +5
Methods for navigation based on large-scale learning typically treat each episode as a new problem, where the agent is spawned with a clean memory in an unknown environment. While…
Zero-BEV: Zero-shot Projection of Any First-Person Modality to BEV Maps
Gianluca Monaci, Leonid Antsfeld, Boris Chidlovskii +1
Bird's-eye view (BEV) maps are an important geometrically structured representation widely used in robotics, in particular self-driving vehicles and terrestrial robots. Existing al…
Space and Time Continuous Physics Simulation From Partial Observations
Janny Steeven, Nadri Madiha, Digne Julie +1
Modern techniques for physical simulations rely on numerical schemes and mesh-refinement methods to address trade-offs between precision and complexity, but these handcrafted solut…
Task-conditioned adaptation of visual features in multi-task policy learning
Pierre Marza, Laetitia Matignon, Olivier Simonin +1
Successfully addressing a wide variety of tasks is a core ability of autonomous agents, requiring flexibly adapting the underlying decision-making strategies and, as we argue in th…
Learning to navigate efficiently and precisely in real environments
Guillaume Bono, Hervé Poirier, Leonid Antsfeld +3
In the context of autonomous navigation of terrestrial robots, the creation of realistic models for agent dynamics and sensing is a widespread habit in the robotics literature and…