Track papers across labs, topics, and trends.
Search papers, labs, and topics across Lattice.
This paper introduces CALF (Convolutional Attention for Leg Features), a novel neural architecture that enables robots to navigate among pedestrians by interpreting leg motion from 2D LiDAR scans. By addressing the limitations of traditional navigation methods that simplify pedestrians to basic shapes, CALF leverages deep reinforcement learning within a custom simulator, LegNav, to produce safe and socially compliant navigation commands. The approach outperforms classical and learning-based baselines and demonstrates effective real-world deployment on a TurtleBot 4, achieving smooth trajectories in pedestrian-rich environments.
Robots can now navigate crowded spaces more effectively by interpreting leg movements rather than treating pedestrians as mere geometric shapes.
Robots navigating among pedestrians typically sense their surroundings with a 2D LiDAR mounted close to the ground. At that height, the sensor mostly sees moving legs rather than whole people, yet most learning-based navigation methods still treat pedestrians as simple shapes like circles. This paper addresses that gap with CALF (Convolutional Attention for Leg Features), an end-to-end neural architecture that combines convolutional layers, attention, and MLP to interpret leg motion directly from LiDAR scans and produce safe navigation commands. The CALF policy is trained using deep reinforcement learning algorithms within LegNav, a custom lightweight 2D simulator that combines 2D LiDAR ray tracing with a novel pedestrian gait model. The resulting policy is compared against classical and learning-based baselines in terms of navigation performance and social compliance. The approach is validated through real-world experiments via zero-shot deployment on a TurtleBot 4, yielding smooth and socially compliant trajectories. Written in JAX, the LegNav simulator enables the training of a deployment-ready CALF policy in under an hour on a single consumer GPU.