CamSLAM: Advanced Real-Time Mapping and Localization

CamSLAM

CamSLAM™ is a navigation software platform that provides fast, accurate real-time mapping and localization. It offers superior navigation, mapping, and safety for a range of robotic and related mobility applications including augmented and virtual reality, person tracking, robot/ drone control, vehicle navigation, and tele-manipulation.


Technology

The CamSLAM software platform exploits highly optimized vision algorithms for accurate localization. It works with multiple sensors for robust, low-power, real-time localization on mobile platforms.

Key Capabilities of CamSLAM Suite

Navigation

  • Precisely locates in 3D both indoors and outdoors
  • Live GPS-denied navigation
  • Superior open-loop navigation algorithms provide high-accuracy estimates of movement with low drift rates

Mapping

  • Visual map generation and map-assisted navigation
  • Maps and tags areas of interest
  • Automatic real-time construction and updating of landmark maps with loop closures to reset navigation drift

Obstacle Detection and Safety

  • Robust stereo vision algorithms
  • Stereo-based 3D sensing enables high-accuracy obstacle detection and avoidance
  • Can use multiple sensors for dynamic performance in challenging conditions including motion blur, bumping into walls, moving people, and dark areas

Path Planning

  • Real-time localization enables repeatable precision
  • Dynamically creates maps to ensure low-drift errors in subsequent runs

Features

  • Can capture and process synchronized data from multiple sensors including accelerometer, gyroscope, magnetometer, barometer, odometer, GPS and monocular/stereo camera data to form an optimal localization solution in real-time
  • Computes six degree of freedom (3D position and 3D orientation) navigation tracks in real time
  • Can continuously track visual features and estimate 3D location of scene points over large windows of time
  • Robust in difficult situations such as crowded areas, dark area, fast or very jerky platform motions
  • Able to generate visual landmark maps on the fly in real time and use landmarks to re-acquire positional constraints when revisiting a site
  • Maps can be stored and reused to improve localization, guide navigation, and provide highly accurate route re-tracing
  • Software libraries run on multiple devices such as those running Windows, Linux, and ARM-based processors
  • Runs on low-powered smartphone processors

Hardware and Sensors

CamSLAM is designed to run on low-cost sensor and processor hardware and can be ported to smartphones. SRI offers clients extensive hardware layer expertise integrating CamSLAM with COTS hardware including SRI sensor-processor boards incorporating multiple sensors: stereo/monocular cameras, IMU, barometer, magnetometer, GPS etc. with embedded FPGA/ARM processor and smartphone processor boards.


Read more from SRI

  • Banner and attendees at the IEEE Hard Tech Venture Summit

    Cultivating hard tech startups that scale

    IEEE’s Hard Tech Venture Summit convened innovators at SRI to refine strategies and build new networks.

  • Patient going into a MRI

    Bringing surgical tools inside the MRI

    Drawing on SRI’s unique innovation ecosystem, the startup Medical Devices Corner is seeking to improve cancer surgery by advancing MRI-safe teleoperation.

  • Christopher Mims and Susan Patrick

    PARC Forum: How to AI

    The Wall Street Journal tech columnist Christopher Mims and SRI Education’s Susan Patrick discuss how AI can strengthen human agency.