Zone-Controlled Machine Safety Using Trusted Object Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing safety systems in industrial environments often fail to detect objects outside the field of view of machine sensors, leading to potential collisions and reduced productivity due to unnecessary emergency stoppages.
Innovation Solution
A method that utilizes fixed and mobile optical sensors to capture images of the operating field, detect objects-of-interest, calculate their positions in a reference coordinate system, and define controlled zones to trigger proactive safety measures such as reducing machine performance or activating warnings, based on trust scores derived from sensor characteristics.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If machine sensors are used to detect objects in the operating field, then collision safety is improved, but objects outside the field of view cannot be detected leading to reduced productivity
Solution Approach 1:
A zone controller acts as an intermediary between fixed optical sensors and the machine. The zone controller receives image data from fixed sensors, detects objects in the machine's blind spots, calculates their positions in the reference coordinate system, and triggers safety measures before the machine reaches hazardous zones. This intermediary system extends detection capability beyond the machine's own sensors without compromising productivity.
Solution Approach 2:
The system performs preliminary safety checks by continuously monitoring the operating field with fixed optical sensors before the machine arrives at potential hazard zones. Objects are detected and tracked in advance, and the zone controller proactively triggers safety measures (reduced performance mode or emergency stop) before the machine reaches the controlled zone, preventing collisions while avoiding unnecessary stoppages.
2Measurement precision
If fixed optical sensors are deployed to expand detection coverage, then detection capability is improved, but system complexity increases
Solution Approach 1:
The zone controller serves multiple functions: it receives and processes image data from fixed optical sensors, performs object detection, calculates object positions in the reference coordinate system, determines when objects enter controlled zones, and triggers appropriate safety measures. This multi-functional approach consolidates complexity into a single controller rather than distributing it across multiple independent systems.
Solution Approach 2:
The zone controller acts as an intermediary layer between the fixed optical sensors and the machine control system. It processes sensor data, makes safety decisions, and communicates with the machine, thereby simplifying the overall system architecture by centralizing the complexity of coordinating multiple sensors with machine control.
3Reliability
If trust score verification is implemented for sensor data, then safety reliability is improved, but processing time increases
Solution Approach 1:
The zone controller continuously monitors and validates sensor data streams in advance, building trust scores over time as objects are tracked across multiple frames. This preliminary verification allows the system to make rapid safety decisions when objects enter controlled zones, as the trust assessment has already been performed during the object's approach.
Solution Approach 2:
The system implements feedback loops where sensor data is continuously validated and trust scores are updated based on object detection consistency, position calculation accuracy, and adherence to expected motion patterns. This feedback mechanism ensures high safety reliability while maintaining efficient processing through adaptive validation thresholds.
Data Source
AI summary
A method includes: accessing message including a frame captured by a fixed sensor in an operating field; detecting an object in the frame; calculating a first position of the object, in a reference coordinate system, within the operating field; calculating a trust score for the object based on a set of characteristics of the message; accessing a second position of a machine in the reference coordinate system; defining a controlled zone, in the reference coordinate system, encompassing the second position occupied by the machine based on the trust score; in response to detecting the first position of the first object intersecting the controlled zone, generating a command that transitions the machine from a nominal operating mode to a reduced performance operating mode; and transmitting the command to the machine.