Grounded curriculum learning
The high cost of real-world data for robotics Reinforcement Learning (RL) leads to the wide
usage of simulators. Despite extensive work on building better dynamics models for …
usage of simulators. Despite extensive work on building better dynamics models for …
Pietra: Physics-informed evidential learning for traversing out-of-distribution terrain
Self-supervised learning is a powerful approach for developing traversability models for off-
road navigation, but these models often struggle with inputs unseen during training. Existing …
road navigation, but these models often struggle with inputs unseen during training. Existing …
Reinforcement learning for wheeled mobility on vertically challenging terrain
Off-road navigation on vertically challenging ter-rain, involving steep slopes and rugged
boulders, presents significant challenges for wheeled robots both at the planning level to …
boulders, presents significant challenges for wheeled robots both at the planning level to …
Vertiencoder: Self-supervised kinodynamic representation learning on vertically challenging terrain
We present VertiEncoder, a self-supervised representation learning approach for robot
mobility on vertically challenging terrain. Using the same pre-training process, VertiEncoder …
mobility on vertically challenging terrain. Using the same pre-training process, VertiEncoder …
GND: Global Navigation Dataset with Multi-Modal Perception and Multi-Category Traversability in Outdoor Campus Environments
Navigating large-scale outdoor environments requires complex reasoning in terms of
geometric structures, environmental semantics, and terrain characteristics, which are …
geometric structures, environmental semantics, and terrain characteristics, which are …
Traverse the non-traversable: Estimating traversability for wheeled mobility on vertically challenging terrain
Most traversability estimation techniques divide off-road terrain into traversable (eg,
pavement, gravel, and grass) and non-traversable (eg, boulders, vegetation, and ditches) …
pavement, gravel, and grass) and non-traversable (eg, boulders, vegetation, and ditches) …
Verti-Selector: Automatic Curriculum Learning for Wheeled Mobility on Vertically Challenging Terrain
Reinforcement Learning (RL) has the potential to enable extreme off-road mobility by
circumventing complex kinodynamic modeling, planning, and control by simulated end-to …
circumventing complex kinodynamic modeling, planning, and control by simulated end-to …
M2P2: A Multi-Modal Passive Perception Dataset for Off-Road Mobility in Extreme Low-Light Conditions
Long-duration, off-road, autonomous missions require robots to continuously perceive their
surroundings regardless of the ambient lighting conditions. Most existing autonomy systems …
surroundings regardless of the ambient lighting conditions. Most existing autonomy systems …