Access Gate Self-Calibrating Sensor Array for Low-Cost Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing controlled access gates face challenges with high sensitivity and costly sensors that require specific positioning and calibration, leading to increased costs and complexity in installation and maintenance.
Innovation Solution
A controlled access gate system utilizing a series of sensors and an electronic control unit with machine-learning and deep-learning algorithms, capable of autonomously identifying and adapting to various types of gates and barriers, regardless of their position or orientation, allowing for effective detection and regulation of subjects passing through.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If high sensitivity sensors (ToF or 3D-vision) are mounted at significant heights (2.3-2.5 meters) to ensure proper detection coverage, then measurement precision and detection area are improved, but installation complexity and cost increase
Solution Approach 1:
The system performs self-calibration by automatically determining sensor positions and orientations through mutual observation between multiple sensors. The sensors autonomously calculate their own spatial relationships without requiring external calibration equipment or expert intervention, thereby eliminating the need for complex manual installation and calibration procedures while maintaining high detection precision
Solution Approach 2:
The patent employs multiple low-cost sensors that can be installed in various positions and orientations, with each sensor contributing to the overall detection capability. The system universally handles different sensor configurations through AI-based calibration, making the installation process adaptable to different mounting scenarios without requiring specialized high-position mounting infrastructure
2Measurement precision
If high sensitivity sensors are used to ensure accurate detection from various positions, then detection accuracy is improved, but cost increases
Solution Approach 1:
The system replaces expensive high-sensitivity ToF or 3D-vision sensors with multiple low-cost sensors that can be mass-produced. By using cheaper sensors in a multi-sensor array and compensating for individual lower precision through collective AI processing and calibration, the system achieves comparable or superior overall detection accuracy at reduced cost
Solution Approach 2:
The patent combines data from multiple low-sensitivity sensors to achieve the detection accuracy that would otherwise require a single high-sensitivity sensor. The AI algorithm merges and correlates observations from all sensors, creating a composite detection capability that exceeds the sum of individual sensor performances while reducing overall system cost
3Ease of operation
If sensors are mounted in random positions or orientations to simplify installation, then ease of installation is improved, but detection reliability deteriorates
Solution Approach 1:
The system dynamically adapts to any sensor configuration through automated calibration. Rather than requiring fixed, pre-determined sensor positions for reliable operation, the system adjusts its detection algorithms in real-time based on the actual positions and orientations of the installed sensors, thereby maintaining high detection reliability regardless of installation variability
Solution Approach 2:
The sensors perform mutual observation to provide feedback information about their own positions and orientations. This self-generated feedback enables the system to automatically calibrate and adjust detection parameters based on the actual installation configuration, ensuring reliable detection performance even when sensors are mounted in random or non-ideal positions
Data Source
AI summary
A controlled access entrance gate, comprising a frame or structure (11) that defines an entry area (I1), an exit area (I2) and a transit area (P) of a user (U), a actuation unit (13), which allows the passage of one or more subjects (U), an electronic control unit (14) and a plurality of sensors or cameras (15, 16). The sensors or cameras (15, 16), which can also be used as an independent kit and can be associated with any type of passage or area to be controlled, are suitable for detecting the data relating to the distance (DT) between each sensor (15, 16) and each subject (U) present in the entry area (I1) or in the exit area (I2) or in the passage area (P) and the speed, trajectory and tracking parameters of the subject (U). An electronic control unit (14) receives and processes data through interpolation processes and machine-learning and/or deep-learning algorithms, so as to autonomously learn the characteristics of the passage and predict the forms and probabilistic directions of each subject (U) inside the volume corresponding to the entry (I1) and exit (I2) areas and to the transit area (P) of the passage.