Vehicle Sensor Circuit Filtering for Stationary Object Recognition
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current sensor systems in motor vehicles face challenges in distinguishing between stationary road users and landmarks at long distances, particularly in scenarios like traffic jams, due to limited range and ambiguity in Doppler-based radar reflections, leading to incorrect identification of stationary vehicles as infrastructure.
Innovation Solution
A method involving a processor unit in a sensor circuit that filters detection points by categorizing them into surface regions using pattern data, allowing for differentiation between road users and stationary landmarks without shape recognition, and adjusts trajectory planning accordingly, using environmental sensors like radar, ultrasound, or cameras.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If radar Doppler-based detection is used to distinguish moving vehicles, then moving objects can be detected, but stationary vehicles cannot be distinguished from highway infrastructure landmarks
Solution Approach 1:
The detection region is segmented into multiple surface regions (road surface region, sidewalk region, median region) based on map data. Each region is evaluated independently to determine whether detection points belong to stationary vehicles or infrastructure landmarks, enabling reliable distinction where traditional Doppler methods fail.
Solution Approach 2:
The system transitions from one-dimensional Doppler velocity-based detection to multi-dimensional spatial analysis by dividing the detection region into multiple surface regions with different categories. This dimensional expansion allows stationary objects to be distinguished based on their spatial location and surface region classification rather than velocity.
2Difficulty of detecting and measuring
If individual detection points are interconnected into objects for recognition, then object detection is enabled, but stationary vehicles are misidentified as highway infrastructure
Solution Approach 1:
The system performs preliminary evaluation of each detection point by determining its associated surface region category before interconnecting detection points into objects. This preliminary classification prevents misidentification by ensuring that detection points from stationary vehicles in road surface regions are distinguished from infrastructure points in sidewalk or median regions before object formation.
3Measurement precision
If computational resources are allocated for detailed object recognition, then accurate identification is achieved, but computational workload increases
Solution Approach 1:
The system applies partial action by performing surface region determination and category evaluation only for detection points that require differentiation between stationary vehicles and infrastructure. Not all detection points undergo the full recognition process, reducing computational workload while maintaining identification precision where needed.
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
Enables accurate recognition of stationary road users from a distance, reducing computational workload and preventing false detections, allowing for timely adjustments in driving trajectory to avoid collisions, even before detailed object recognition is performed.
Implementation Method 1
an object in an environment of the motor vehicle reflects a signal, as is known, for example electromagnetic radiation or ultrasound, and this reflection signal can be received by way of an environmental sensor
Implementation Method 2
the temporarily stationary objects (vehicles) at the end of a traffic jam have the same speed as the highway infrastructure (that is, landmarks such as crash barriers, tunnel entrances, sign gantries), one of the strong points of radar, namely, to distinguish radar reflections in the Doppler range (based on the speed of movement of the other vehicles)
Data Source
AI summary
The disclosure relates to a method for operating a sensor circuit in a motor vehicle, wherein a processor in the sensor circuit receives a reflection signal from an environment of the motor vehicle by way of an environmental sensor and ascertains from the reflection signal a particular detection point, which describes a relative position of a reflection event with respect to the motor vehicle. The disclosure involves the processor using pattern data to establish a surface region in the environment and a category of the surface region and checking for the detection point whether it lies in the surface region, and the detection point which lies in a surface region is categorized according to the pattern data of the surface region and the detection point is filtered according to its categorization by way of a filtering action for trajectory planning of a driving trajectory of the motor vehicle.