RTISSLAM-3D: Advanced 3D Sonar System for Wheeled Robots

Paper: RTISSLAM-3D: Full 6-DoF SLAM With an In-Air 3D Imaging Sonar

Authors: Jan Steckel, Vincent Katter, Wouter Jansen

Executive Summary

Recent research has developed a new system called RTISSLAM-3D, which enhances the navigation capabilities of wheeled robots in complex environments such as dense forests. Traditional sonar systems primarily function in two dimensions, often overlooking important height information, which can be critical when navigating uneven or sloped terrains. This new approach integrates a three-dimensional imaging sonar with additional sensors like an Inertial Measurement Unit (IMU) and wheel encoders, allowing for a more comprehensive understanding of the robot's surroundings.

The RTISSLAM-3D system employs advanced techniques to process sonar data. It uses a Constant False Alarm Rate (CFAR) detector to distill sonar images into a manageable set of three-dimensional points. This 3D data is essential for accurately tracking the robot’s movement in three dimensions, recognizing its position in space, and verifying that it has returned to previously visited locations. By creating a pose graph that includes these various measurements, RTISSLAM-3D can maintain an accurate representation of its route and environment.

The effectiveness of RTISSLAM-3D was tested over a distance of 545 meters in a forest setting, with results benchmarked against a well-established 3D lidar SLAM system. The findings indicated significant improvements in accuracy. The loop closure technique—essentially confirming that the robot had returned to a known location—reduced the horizontal error of the navigation system from 1.74 meters to just 0.28 meters. Additionally, the height measurement error was recorded at 0.12 meters. Remarkably, 94.6% of the 3D sonar map produced by RTISSLAM-3D fell within 0.3 meters of the lidar map, demonstrating a high level of accuracy in the 3D mapping process.

This research may have substantial implications for industries reliant on robotic navigation in challenging environments, such as agriculture, forestry, and search and rescue operations. By providing robots with the ability to accurately navigate and map complex terrains, organizations could improve operational efficiency and safety. While the current findings are based on simulated benchmarks rather than real-world deployments, the insights gained from RTISSLAM-3D could pave the way for further developments in autonomous navigation technologies.

Academic Abstract

Echolocating bats navigate dense forests using sonar alone, and in-air 3D imaging sonars now capture the entire frontal hemisphere with every pulse. Sonar SLAM, however, has so far been planar: current systems estimate a 2D trajectory and discard the elevation information in the sonar images, an assumption that does not hold on rough and sloped terrain such as a forest floor. In this paper, we introduce RTISSLAM-3D, a novel 6-DoF SLAM system for a wheeled robot that carries an in-air 3D imaging sonar, an IMU and wheel encoders. A CFAR detector on the sphere reduces every sonar image to a sparse set of 3D detections, which we use for a 3D acoustic odometry that models the shape of the sonar beam, for place recognition with polar submap descriptors, and for the verification of loop closures by 3D registration, all combined in a 6-DoF pose graph with attitude factors. We evaluated RTISSLAM-3D on a 545m drive through a forest, with a 3D lidar SLAM as reference. The loop closures reduced the horizontal error of the acoustic odometry from 1.74m to 0.28m, with a height error of 0.12m, and 94.6% of the resulting 3D sonar map lies within 0.3 m of the lidar map.