Blind Spot Object Tracking via Motion Model and Intermediary Markers
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing vehicle computer systems face challenges in operating vehicles safely and efficiently when trying to connect to objects in blind spots, as sensors may not provide accurate data due to overlapping fields of view or unreliable data in low light conditions, leading to difficulties in hitching trailers.
Innovation Solution
A system and method using a processor to identify and track objects by creating a motion model based on relationships between points on the object, including dead reckoning and sensor data, to determine the location of points in blind spots, allowing the vehicle to navigate and connect to objects like trailers even when they are outside the sensors' field of view.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Difficulty of detecting and measuring
If vehicle sensors are used to detect objects, then object detection capability is improved, but detection reliability deteriorates in blind spots
Solution Approach 1:
The patent uses visible light markers as intermediaries to bridge the gap between the vehicle and objects in blind spots. These markers reflect light from the vehicle's own illumination system, allowing the sensor to detect objects that would otherwise be invisible in the blind spot area, thus improving detection reliability without adding external sensors.
Solution Approach 2:
The patent creates a visual copy or representation of the object by projecting structured light patterns onto markers attached to the object. The sensor captures the reflected light pattern, which serves as a copy of the object's position and orientation information, enabling accurate detection in blind spots where direct sensing would fail.
2Area of stationary object
If sensor fields of view are expanded to cover blind spots, then detection coverage is improved, but measurement precision deteriorates due to overlapping fields of view
Solution Approach 1:
The patent divides the detection task into segments by using multiple sensors with overlapping fields of view, each responsible for detecting markers in specific angular ranges. The system processes data from each sensor segment independently and then combines the results, maintaining high precision by avoiding the need to process ambiguous overlapping data as a single unit.
Solution Approach 2:
The patent applies different processing strategies to different spatial regions. In areas where sensor fields overlap, the system uses geometric algorithms to resolve ambiguities by selecting the most reliable sensor data based on signal strength and angular position. This local optimization maintains measurement precision across the entire expanded coverage area.
3Device complexity
If traditional sensor-only methods are used, then system complexity is kept low, but hitching efficiency deteriorates due to inability to track objects in blind spots
Solution Approach 1:
The patent makes the vehicle's existing illumination system serve a dual purpose: both lighting for the driver and active illumination for the sensor to detect markers on objects in blind spots. This self-service approach enables blind spot detection without adding dedicated illumination devices, maintaining low system complexity while improving hitching efficiency through accurate object tracking.
Data Source
AI summary
A system includes a processor and a memory. The memory stores instructions executable by the processor to identify an object, including points on the object, to perform tracking of the object based on a motion model including a relationship of the points to one another, and to then output a location of one of the points in a blind spot based on the tracking.