Surveillance Robot Actions for Resolving Facility State Ambiguity

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current surveillance systems for facilities lack effective methods to dynamically evaluate and classify anomalies in a facility's state, particularly in complex environments with varying conditions, leading to high false alarm rates and limited adaptability to specific site characteristics.

Innovation Solution

A facility surveillance system utilizing a dynamic building information model (BIM) with a central computing unit and multiple sensors, including cameras, motion detectors, and contact sensors, that generates and analyzes surveillance data to classify states as normal or anomalous based on topological, logical, and functional relationships, using machine learning and feedback mechanisms to improve classification accuracy.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If traditional surveillance systems are used to monitor facility elements, then basic surveillance coverage is achieved, but false alarm rates are high and adaptability to site characteristics is limited

Engineering Contradiction:
Improveclassification accuracyVSAvoidadaptability to site characteristics
Core Design Contradiction:
ReliabilityVSAdaptability or versatility

Solution Approach 1:

The system implements a dynamic building information model that continuously updates facility element states and relationships based on surveillance data. The classification model adapts its parameters and thresholds dynamically according to learned site characteristics, transforming the static surveillance system into a dynamic one that improves reliability while maintaining adaptability.

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The system incorporates feedback mechanisms where classification results and surveillance data are fed back into the building information model to refine site characteristics. This feedback loop enables the system to learn from operational data, improving classification accuracy over time while adapting to specific facility environments without requiring manual reconfiguration.

Inventive Principle:
Principle #23Feedback

2Measurement precision

If multiple sensors and complex analysis methods are deployed to improve anomaly detection, then detection capability is enhanced, but system complexity increases

Engineering Contradiction:
Improveanomaly detection capabilityVSAvoidsystem complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The system segments the facility into discrete building information model elements with defined relationships. Each facility element can be monitored independently with specific sensors, and anomalies are detected by analyzing deviations from expected element states and relationships. This segmentation enables precise anomaly detection while maintaining manageable system complexity through modular architecture.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The building information model serves as a universal framework that integrates multiple sensor types and analysis methods. The same BIM structure supports various surveillance functions including state monitoring, relationship validation, and anomaly classification, reducing overall system complexity by providing a multi-functional platform rather than separate specialized systems.

Inventive Principle:
Principle #6Universality (Multi-functionality)

3Speed

If real-time surveillance and classification are performed continuously, then responsive anomaly detection is achieved, but computational resource consumption increases

Engineering Contradiction:
Improveresponse timeVSAvoidcomputational resource consumption
Core Design Contradiction:
SpeedVSUse of energy by moving object

Solution Approach 1:

The system performs partial classification by focusing computational resources on evaluating only those facility element states that deviate from expected conditions or show anomaly indicators. Rather than continuously classifying all elements at full detail, the system applies classification selectively to suspicious states, achieving responsive anomaly detection while reducing overall computational resource consumption.

Inventive Principle:
Principle #16Partial or excessive action

Data Source

PatentEP3989194A1Facility surveillance systems and methods
Publication Date: 2022.04.27 HEXAGON INNOVATION HUB GMBH
  • EP3989194A1 patent drawingFigure 1~2a
  • EP3989194A1 patent drawingFigure 2b~2c
  • EP3989194A1 patent drawingFigure 2d~3

AI summary

The invention pertains to a surveillance system (103) for surveillance of a facility (102), the system (103) comprising a surveillance robot (100) having an action controller for controlling actions of the robot (100). Furthermore, the system (103) also comprises • at least a first surveillance sensor (110) designed to acquire first surveillance data of at least one object (105-108, 150-152) of the facility (102), and • a state detector (100c) configured to detect at least one state (114) associated with the object (105-108, 150-152) based on the surveillance data. According to the invention, the state detector (100c) is configured to • notice a state ambiguity (115) of the state (114), in particular by comparison with a predefined ambiguity threshold, • trigger an action (117, 120, 130) of the robot (100) by the action controller in case a state ambiguity is noticed, the triggered action (117, 120, 130) adapted to generate state verification information about the object (105-108, 150-152), the state verification information being suitable to resolve the state ambiguity, and • resolve the state ambiguity considering the state verification information.