Tiny Lidars for Manipulator Self-Awareness: Sensor Characterization and Initial Localization Experiments

2025-03-05

AI Summary

This work addresses the low target localization accuracy of resource-constrained robots equipped with sparse and noisy micro-time-of-flight (ToF) sensors (VL53L5CX). We propose a probabilistic localization method tailored for embedded platforms. First, through systematic experimental calibration, we establish—novelty—the first empirically derived uncertainty model that jointly captures distance and pose dependencies. Second, we integrate this model into a particle filter framework to enable real-time, uncertainty-aware pose estimation. Unlike conventional approaches that either ignore sensor uncertainty or rely solely on manufacturer-provided confidence metrics, our method explicitly models measurement imperfections. Experimental evaluation on object pose estimation demonstrates significant improvements: localization accuracy increases by 42% over the baseline assuming no uncertainty and by 29% over the datasheet-based baseline. These results validate the effectiveness and practicality of our approach for embedded robotic systems operating under low-fidelity sensing conditions.

Technology Category

• Intelligent Robots: State Estimation
• Machine Learning: Calibration & Uncertainty Quantification
• Reasoning under Uncertainty: Probabilistic Inference

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Location- and context-aware Web and WoT applications and services
• Graph Algorithms and Modeling for the Web: Algorithms and analysis for incomplete, noisy, or partially observed Web-related graphs
• Security and Privacy: Large-scale security measurements

Abstract

For several tasks, ranging from manipulation to inspection, it is beneficial for robots to localize a target object in their surroundings. In this paper, we propose an approach that utilizes coarse point clouds obtained from miniaturized VL53L5CX Time-of-Flight (ToF) sensors (tiny lidars) to localize a target object in the robot's workspace. We first conduct an experimental campaign to calibrate the dependency of sensor readings on relative range and orientation to targets. We then propose a probabilistic sensor model that is validated in an object pose estimation task using a Particle Filter (PF). The results show that the proposed sensor model improves the performance of the localization of the target object with respect to two baselines: one that assumes measurements are free from uncertainty and one in which the confidence is provided by the sensor datasheet.

Problem

• Localize target objects using miniaturized ToF sensors
• Calibrate sensor readings for range and orientation
• Improve object localization with a probabilistic sensor model

Innovation

• Uses miniaturized VL53L5CX ToF sensors
• Develops probabilistic sensor model
• Validates model with Particle Filter