Railway Signalling Layout Checking With Bounded Graph Search
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The verification of railway signalling designs is labor-intensive, prone to human error, and often incomplete due to the complexity and safety-critical nature of the systems, necessitating a more efficient and exhaustive testing method.
Innovation Solution
A computer-implemented method using a graph representation and bounded search with incremental variable bounds to check railway signalling designs, employing an automatic theorem prover to simulate a breadth-first search and verify the geographical positions of railway assets against predefined rules and axioms.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If manual verification by trained operators is used, then the verification can be performed with human judgment and flexibility, but the process becomes labor-intensive and prone to human error
Solution Approach 1:
The patent replaces the mechanical system of manual verification by trained operators with an automated computer-implemented verification system. The system uses a processor to automatically check railway signalling designs against stored rules and axioms, eliminating human labor while maintaining verification accuracy through systematic automated reasoning.
Solution Approach 2:
The verification system performs self-service by automatically checking its own design outputs against predefined rules and axioms without requiring external human operators. The processor independently validates the signalling scheme by executing logical checks and generating verification results autonomously.
2Reliability
If exhaustive testing of all possible states is performed, then complete verification coverage is achieved, but the testing becomes non-feasible due to the complexity of railway signalling systems
Solution Approach 1:
The patent segments the complex verification problem into manageable components by dividing the railway signalling design into discrete elements (signals, track sections, routes) and verifying each against specific rules and axioms. This segmentation makes exhaustive verification feasible by breaking down the overwhelming complexity into systematic, checkable units.
Solution Approach 2:
The system performs preliminary action by pre-storing rules and axioms that define correct signalling behavior before verification begins. These predefined logical constraints enable the processor to systematically check all possible states without requiring complex real-time analysis during the verification process.
3Ease of manufacture
If multiple systems and software components are tested individually, then each component can be verified separately, but the overall verification process becomes labor-intensive and costly
Solution Approach 1:
The patent merges the verification of multiple separate systems and software components into a single integrated verification process. The processor simultaneously checks all signalling elements, software components, and their interactions against the stored rules and axioms, reducing the overall verification time while maintaining the ability to verify individual components when needed.
Data Source
Figure 1~3
Figure 4~5
AI summary
A computer-implemented method of checking the design of a railway signalling scheme is described. A track plan is rendered as an attributed graph, with a set of rules and axioms in first order logic defining the behaviour of the graph, along with annotations on the graph, used to determine whether or not the geographical position of an individual railway asset on the track plan is correct. Each rule corresponds to the correct geographical position of an individual railway asset. For each individual railway asset, a bounded search of the annotated graph using a pre-determined set of bounds to check whether a rule corresponding to the correct geographical position of the individual railway asset holds true. The pre-determined set of bounds contains a variable bound k that is incrementally increased until the result of the rule check is known.