Smarter and Thinner Self-navigating Robots Using Hybrid Time-of-Flight

Modern robotic vacuum cleaners (RVC) utilise simultaneous localization and mapping (SLAM) to navigate their environment. In addition, users are provided with a floorplan, which can be used for smartphone app-based control to define cleaning and restricted areas. REAL3™ hybrid Time-of-Flight (hToF) combines a single ToF image sensor in a wide 100° FoV setup with two illumination types to offer an improvement to currently prevailing ranging technologies in RVCs.

• Spot grid illumination word providing long-range 3D point cloud data for SLAM with a range of up to 8-10m.
• Homogeneous flood illumination enables obstacle avoidance via close- to mid-range point cloud data with high lateral resolution with approx. 21600 depth points and a range up to 3m.

Hybrid Time-of-Flight technology delivers a high-quality, cost-efficient solution for SLAM and obstacle avoidance, reducing size and system complexity compared to standard LDS (laser distance sensor) towers, while offering substantial advantages over stereo vision or structured light.

The webinar will introduce SLAM for consumer robots using hToF. The achievable accuracy in localization and mapping will be discussed as well as heuristics in parameter optimization using Google Cartographer. This will be presented in the context of processing benchmarks for different embedded SoC platforms using similar processor cores as frequently used in consumer robots.

The unique hToF technology is already adopted by one of the leading robot manufacturers, for instance the recent Roborock V20 launch in March 2024.