3D Point Cloud Masking for AGV No-Entry Zone Control

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing systems for preventing autonomous guided vehicles from entering specific locations require advanced expertise, high-performance calculation devices, and increased power consumption, and are hindered by unreliable distance measurement at object-background boundaries.

Innovation Solution

A mask region setting unit sets a region on 3D point clouds from a distance measuring sensor based on intensity images, allowing smoothing filter processors to output mask regions without smoothing, thereby using 3D point clouds to create virtual barricades around specific locations using characteristic shapes.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of operation

If image recognition is used to prevent autonomous guided vehicles from entering specific locations, then control functionality is achieved, but prior data acquisition and machine learning are required, advanced expertise is needed, high-performance calculation devices are required, and power consumption increases

Engineering Contradiction:

Improveease of implementationVSAvoidsystem complexity

Core Design Contradiction:

Ease of operationVSDevice complexity

Solution Approach 1:

The patent replaces expensive, complex image recognition systems with a simple, inexpensive reflective tape that can be easily attached and removed. The tape serves as a temporary, disposable marker that creates reliable depth detection signals without requiring advanced processing systems, expertise, or high power consumption.

Inventive Principle:

Principle #27Cheap short-living objects (Disposable)

Solution Approach 2:

The patent substitutes optical image recognition processing with a simpler depth measurement system using time-of-flight sensors. By reflecting light off the tape and measuring phase differences, the system achieves reliable location identification through physical light measurement rather than complex image analysis, reducing computational requirements.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

2Reliability

If smoothing processing is applied to 3D point clouds, then measurement reliability improves, but depth detection accuracy at object boundaries deteriorates

Engineering Contradiction:

Improvemeasurement reliabilityVSAvoiddepth detection accuracy

Core Design Contradiction:

ReliabilityVSMeasurement precision

Solution Approach 1:

The patent applies different processing treatments to different regions of the 3D point cloud based on detected characteristic shapes. When reflective tape is detected, the system creates a mask region that preserves raw depth data without smoothing, maintaining accuracy at the object boundary while allowing smoothing in other regions to improve overall measurement reliability.

Inventive Principle:

Principle #3Local quality

Solution Approach 2:

The patent introduces a mask region as an intermediary layer between the raw 3D point cloud and the final processed data. This mask selectively protects depth values in the vicinity of characteristic shapes from smoothing processing, allowing the system to maintain both reliability through smoothing and precision at boundaries through selective preservation.

Inventive Principle:

Principle #24Intermediary (Mediator)

Applied Scientific Principles

This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.

Function Achieved in This Case

This approach allows easy and cost-effective control of autonomous guided vehicles by preventing entry into specific areas without the need for prior data acquisition, machine learning, or high-performance devices, reducing power consumption and avoiding erroneous depth detections.

Implementation Method 1

a distance measuring sensor that measures a distance by accumulating reflected light and detecting a phase difference between irradiation light and the reflected light

Methodology Applied

Scientific EffectPhase difference detection:

Data Source

PatentUS20260030726A1Information processing apparatus, information processing method, and information processing system

Publication Date: 2026.01.29 SONY GROUP CORP

AI summary

Control for preventing a mobile body such as an autonomous guided vehicle from entering a specific location is achieved easily and inexpensively.A mask region setting unit sets a mask region not to be subjected to smoothing on a 3D point cloud output from a distance measuring sensor on the basis of an intensity image output from the distance measuring sensor that measures a distance by accumulating reflected light and detecting a phase difference between irradiation light and the reflected light. A smoothing filter processor performs smoothing processing on the 3D point cloud output from the distance measuring sensor. In this case, the 3D point cloud of the set mask region is output as it is without being subjected to the smoothing processing. In the 3D point cloud after smoothing filter processing, it is possible to leave the 3D point cloud related to erroneous depth detection only in a necessary region, and by using the 3D point cloud, control for preventing a mobile body such as, for example, an autonomous guided vehicle from entering a specific location can be achieved easily and inexpensively.