Reflective Surface Object Detection Beyond Sensor Field of View
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing vehicle sensing systems struggle to accurately track objects outside their normal field of view due to limitations in sensor range, leading to potential collisions by failing to detect objects in reflective surfaces such as mirrors, windows, and puddles.
Innovation Solution
Utilizing a combination of neural networks and Kalman filters to detect and track reflective surfaces, determine their real-world locations, and predict object trajectories based on sensor attributes and geometric properties, enabling the calculation of probabilities of contact with objects outside the sensor's direct line of sight.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the sensor field of view is expanded to cover more area, then the ability to detect objects increases, but the sensor complexity and cost increase
Solution Approach 1:
The patent uses reflective surfaces (mirrors, windows, puddles) as intermediaries to extend the effective field of view. These surfaces act as mediators that reflect light from objects outside the direct sensor view, allowing the sensor to detect objects indirectly without requiring additional sensors or expanding the physical sensor array.
Solution Approach 2:
The patent transitions from direct line-of-sight detection (one dimension) to indirect detection via reflective surfaces (adding another dimension). By detecting reflections and calculating virtual camera positions, the system effectively adds a dimensional aspect to object detection, enabling detection of objects that would otherwise be outside the direct field of view.
2Reliability
If the sensor range is increased to detect distant objects, then collision avoidance improves, but the sensor precision and calibration requirements worsen
Solution Approach 1:
The patent creates virtual copies of the camera at reflected positions based on reflective surface geometry. By calculating where virtual cameras would be positioned relative to reflective surfaces, the system can determine object locations with precision equivalent to having physical sensors at those positions, without actually placing sensors there.
Solution Approach 2:
The system performs preliminary detection by identifying reflective surfaces and calculating virtual camera positions before actual object detection occurs. This pre-computation of geometric relationships enables precise object location determination from reflected light paths.
3Reliability
If objects outside the normal field of view are detected using reflective surfaces, then safety improves, but the computational complexity increases
Solution Approach 1:
The patent segments the detection problem into distinct components: identifying reflective surfaces, calculating virtual camera positions for each surface, detecting objects in reflected views, and computing object locations based on virtual camera geometry. This segmentation allows the computational task to be broken into manageable steps that can be processed efficiently.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
Enhances the ability to predict potential collisions with objects not directly visible to sensors, allowing vehicles to adjust trajectories proactively and avoid contact, thereby improving safety and operational efficiency.
Implementation Method 1
detecting a reflective surface in a plurality of images acquired by a sensor included in the vehicle
Data Source
AI summary
Tracking an object included in a reflective surface by detecting the reflective surface included in one or more images included in a plurality of images by determining a location of the reflective surface in pixel coordinates and tracking the location of the reflective surface in the plurality of images. Real world locations of the object can be determined based on attributes of the reflective surface including a geometric class, extrinsic properties that relate the reflective surface to an environment, intrinsic properties describing the reflective surface without referring to the environment and calibration properties of the sensor.