Estimating Scene Flow in Robot Surroundings with Distributed Miniaturized Time-of-Flight Sensors

2025-04-03

AI Summary

To address the need for safe physical human–robot interaction, this paper investigates real-time scene flow estimation from low-density, high-noise point clouds acquired by distributed miniature time-of-flight (ToF) sensors mounted on a robot’s body. We propose a dense motion estimation framework integrating clustering with an enhanced iterative closest point (ICP) algorithm. Our approach introduces a novel fitness-driven mechanism for classifying points as static or dynamic, coupled with an adaptive inlier rejection strategy to mitigate noise and sparsity effects. Furthermore, we incorporate match-quality-based dynamic thresholding and robust geometric correspondence optimization. Evaluated on a 24-sensor experimental platform, the method achieves motion direction and velocity estimation errors comparable to the intrinsic noise level of the ToF sensors. The framework demonstrates strong practicality and robustness under realistic operating conditions.

Technology Category

• Intelligent Robots: State Estimation
• Computer Vision: Motion & Tracking
• Planning, Routing, and Scheduling: Activity and Plan Recognition

Application Category

• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Systems and Infrastructure for Web, Mobile and WoT: Experiences and lessons learnt from Web-based algorithms and system deployments
• Security and Privacy: Large-scale security measurements

Abstract

Tracking motions of humans or objects in the surroundings of the robot is essential to improve safe robot motions and reactions. In this work, we present an approach for scene flow estimation from low-density and noisy point clouds acquired from miniaturized Time of Flight (ToF) sensors distributed on the robot body. The proposed method clusters points from consecutive frames and applies Iterative Closest Point (ICP) to estimate a dense motion flow, with additional steps introduced to mitigate the impact of sensor noise and low-density data points. Specifically, we employ a fitness-based classification to distinguish between stationary and moving points and an inlier removal strategy to refine geometric correspondences. The proposed approach is validated in an experimental setup where 24 ToF are used to estimate the velocity of an object moving at different controlled speeds. Experimental results show that the method consistently approximates the direction of the motion and its magnitude with an error which is in line with sensor noise.

Problem

• Estimating scene flow in robot surroundings using distributed ToF sensors
• Tracking human and object motions for safer robot interactions
• Overcoming noise and low-density data in point cloud motion estimation

Innovation

• Distributed miniaturized Time-of-Flight sensors
• Iterative Closest Point for dense motion flow
• Fitness-based classification and inlier removal