Object Detection System Using Pre-stored Infrastructure Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current vehicle object detection systems face challenges in achieving high-integrity detection with reduced complexity, processing overhead, and characterization uncertainty, particularly in distinguishing infrastructure and non-infrastructure objects while determining vehicle paths and trajectories.
Innovation Solution
A system and method that combine on-board stored information with real-time data using sensors, position and motion devices, and a logic unit to detect objects by determining vehicle paths and trajectories, and using sensor data to differentiate between infrastructure and non-infrastructure objects, thereby reducing complexity and uncertainty.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional sensor-based object detection systems are used, then object detection capability is provided, but system complexity and processing overhead increase
Solution Approach 1:
The system pre-stores infrastructure information (track geometry, infrastructure locations, sightlines) before detection operations. This preliminary preparation allows the detection algorithm to focus only on identifying deviations from the stored model, rather than processing all environmental data from scratch, thereby reducing real-time computational complexity while maintaining high detection integrity
Solution Approach 2:
The detection problem is segmented into two distinct parts: infrastructure objects (detected by comparing sensor data against stored infrastructure information) and non-infrastructure objects (detected by identifying deviations from the expected infrastructure model). This segmentation allows each type to be processed with optimized algorithms, reducing overall system complexity
2Reliability
If comprehensive sensor data processing is performed to detect all objects, then detection coverage is improved, but processing overhead increases
Solution Approach 1:
The system extracts and stores infrastructure-specific information (track geometry, signal locations, platform positions) separately from general environmental data. During detection, only relevant portions of stored information are retrieved and compared with current sensor data, extracting only the necessary processing steps while eliminating redundant computations, thus improving processing efficiency without compromising detection coverage
Solution Approach 2:
By pre-processing and storing infrastructure information in structured formats with defined sightlines and detection zones, the system prepares detection templates in advance. This allows real-time processing to focus only on comparing current sensor readings against pre-defined expectations, significantly reducing processing overhead while maintaining comprehensive detection coverage
3Reliability
If sensor data is used to detect both infrastructure and non-infrastructure objects, then detection capability is enhanced, but characterization uncertainty increases
Solution Approach 1:
The system segments object detection into two distinct categories with different processing approaches: infrastructure objects are identified by matching sensor data against stored infrastructure information (high precision through model comparison), while non-infrastructure objects are detected as deviations from the expected infrastructure model (enhanced capability through anomaly detection). This segmentation reduces characterization uncertainty by applying appropriate methods to each object type
Solution Approach 2:
The stored infrastructure information serves as an intermediary reference model that mediates between raw sensor data and object characterization. By comparing current sensor readings against this stable, pre-characterized infrastructure model, the system achieves precise characterization of infrastructure objects while simultaneously identifying non-infrastructure objects as deviations, thereby reducing overall characterization uncertainty
Data Source
AI summary
A method and system include receiving positioning and motion information from one or more positioning and motion devices on a vehicle on a guideway, based on the positioning and motion information, receiving information from a database on the vehicle, the information comprising track geometry information and infrastructure information corresponding to the guideway, using the track geometry information and positioning and motion information to determine a path of the guideway and a trajectory of the vehicle along the path, receiving data from one or more electromagnetic sensors on the vehicle, and detecting an object by using the trajectory, the infrastructure information, and the information from the one or more electromagnetic sensors.