Memory aware multi sensor navigation improves safety and efficiency
Memory-Aware Multi-Sensor Perception for Efficient and Safe Navigation in Dynamic Environments
Robotics
Summary
Navigating safely in new and changing places is hard because a robot needs to remember what it has seen to avoid obstacles and find its way. The authors created a system that uses both laser and camera sensors combined with a memory that keeps track of the environment as it moves around. This system helps the robot remember where walls and moving objects are, so it can plan better routes and avoid collisions. Their tests show the robot reaches its goals faster and safer, even in tricky spots like narrow halls or busy areas.
What this means in practice
- •For autonomous mobile robot developers: Improve onboard perception and navigation in uncertain environments by integrating multi-sensor memory to handle moving obstacles and narrow pathways safely.
- •For warehouse automation teams: Enable robots to safely navigate bustling indoor facilities by persistently tracking static infrastructure and dynamic objects to reduce collisions and improve efficiency.
Authors
Jingshuo Li, Yifan Xue, Yifei Li, Shubhodeep Shiv Aditya, Nadia Figueroa
Abstract
Autonomous navigation in previously unseen environments requires effective perception, persistent environmental representation, and collision avoidance while maintaining progress toward a goal. Existing perception-based methods often rely on prior maps or short-horizon observations, limiting their ability to exploit previously observed structure. We propose a memory-aware multi-sensor navigation framework that integrates LiDAR and RGB perception, online distance-field representation learning, and a stage-adaptive Modulated Control Barrier Function Quadratic Program (MCBF-QP). The framework persistently represents static infrastructure while tracking dynamic obstacles, enabling the MCBF-QP controller to exploit previously observed geometry for obstacle circumvention and adapt its safety constraints and guidance to local conditions. Experiments in complex indoor and outdoor environments demonstrate improved navigation efficiency and goal-reaching performance while maintaining collision avoidance in narrow passages and around dynamic obstacles.