6 papers
LeCropFollow: Latent Space Planning for Navigation in Unstructured Crop Fields
Felipe Tommaselli, Francisco Affonso, Arthur Pompeu +4
Unstructured navigational features, such as irregular planting or discontinuities, remain the primary failure mode for under-canopy agricultural robots. Existing geometric approach…
CTS-MoE: Implicit Terrain Adaptation via Mixture-of-Experts for Perceptive Locomotion
Francisco Affonso, Matheus P. Angarola, Ana Luiza Mineiro +3
Perceptive legged locomotion over discontinuous terrain (e.g., stairs, gaps, and obstacles) requires adaptive behavior, as a single conservative gait cannot produce the anticipator…
CATNAV: Cached Vision-Language Traversability for Efficient Zero-Shot Robot Navigation
Aditya Potnis, Francisco Affonso, Shreya Gummadi +2
Navigating unstructured environments requires assessing traversal risk relative to a robot's physical capabilities, a challenge that varies across embodiments. We present CATNAV, a…
Learning Terrain-Specialized Policies for Adaptive Locomotion in Challenging Environments
Matheus P. Angarola, Francisco Affonso, Marcelo Becker
Legged robots must exhibit robust and agile locomotion across diverse, unstructured terrains, a challenge exacerbated under blind locomotion settings where terrain information is u…
End-to-End Crop Row Navigation via LiDAR-Based Deep Reinforcement Learning
Ana Luiza Mineiro, Francisco Affonso, Marcelo Becker
Reliable navigation in under-canopy agricultural environments remains a challenge due to GNSS unreliability, cluttered rows, and variable lighting. To address these limitations, we…
Learning to Walk With Less: A Dyna-Style Approach to Quadrupedal Locomotion
Francisco Affonso, Felipe Tommaselli, Felipe Andrade G. Tommaselli +6
Traditional on-policy reinforcement learning (RL) controllers for quadrupedal locomotion often suffer from low data efficiency, requiring millions of interactions with simulated en…