Kudan announced that Kudan's Visual SLAM technology is integrated with NVIDIA Isaac Perceptor, an Isaac ROS 2 software package for developing autonomous mobile robots. NVIDIA Isaac Perceptor is a collection of hardware-accelerated packages for visual AI, tailored for Autonomous Mobile Robot to perceive, localize, and operate robustly in unstructured environments. With this collaboration, Kudan's CUDA-accelerated VSLAM significantly enhances its performance by leveraging the 3D perception functions of NVIDIA Isaac Perceptor, including AI-based depth perception and real-time 3D reconstruction with a cost map.

Using outputs from Kudan VSLAM, NVIDIA Isaac Perceptor can elevate localization performance in environments with dynamic objects and frequent scenery changes. This has been successfully demonstrated using NVIDIA's Nova Carter robot, confirming the value of real-time 3D occupancy maps in enhancing the VSLAM approach for mapping and localization of mobile robots. The combination of Kudan VSLAM with NVIDIA Isaac Perceptor offers a sophisticated, visual-based perception and navigation solution for AMR developers, helping enable reliable localization, obstacle detection, and navigation capabilities in complex scenarios.

Kudan will continue to deliver top-notch, best-in-class perception and navigation solutions, addressing the challenges faced by AMR developers and accelerating the deployment of AMRs in various environments.