Ultrasonic Surrounding Map Fusion for False Obstacle Correction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Ultrasonic sensors in parking assistance systems face challenges with noise in sensor data, leading to inaccurate obstacle positioning due to false targets and ghost objects, which affect the reliability of surrounding maps.
Innovation Solution
An apparatus and method using multiple ultrasonic sensors to generate a surrounding map by employing a processor to create and correct fusion masks based on sensor data, combining current and predicted masks to remove noise and improve accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If ultrasonic sensor data is used to estimate obstacle position, then obstacle detection is enabled, but noise in sensor data causes incorrect position estimation
Solution Approach 1:
The patent combines data from multiple ultrasonic sensors (first ultrasonic sensor and second ultrasonic sensor) to form a comprehensive surrounding map. By merging the detection results from multiple sensors, the system can cross-validate information and reduce the impact of noise or false detections from individual sensors, thereby improving both reliability and precision of obstacle position estimation.
Solution Approach 2:
The patent implements a feedback mechanism where the processor continuously updates the surrounding map by comparing current sensor data with previously stored map information. When noise or incorrect data is detected, the system uses feedback to correct the map, ensuring that only reliable obstacle position information is maintained. This feedback loop allows the system to learn from previous detections and refine future position estimations.
2Measurement precision
If multiple ultrasonic sensors are used to improve detection accuracy, then noise removal is enhanced, but device complexity increases
Solution Approach 1:
The processor performs multiple functions using a single computational unit: it processes data from multiple ultrasonic sensors, generates the surrounding map, identifies obstacles, and updates the map in real-time. By making the processor multi-functional, the patent avoids adding separate dedicated hardware for each function, thereby improving measurement precision through multi-sensor fusion without proportionally increasing device complexity.
Solution Approach 2:
The patent segments the processing task into distinct functional modules within the processor: data collection from sensors, map generation, obstacle identification, and map updating. This segmentation allows each function to be optimized independently while being managed by a single processor, reducing overall system complexity compared to having separate dedicated systems for each function.
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
The method enhances the accuracy of obstacle detection by reducing false targets and improving the reliability of surrounding maps, ensuring precise positioning of objects around a moving vehicle.
Implementation Method 1
The ultrasonic sensor periodically transmits high-frequency sound waves (ultrasonic waves). When the transmitted ultrasonic signal collides with (hits) an obstacle, the ultrasonic sensor receives a signal reflected from the obstacle.
Implementation Method 2
When the transmitted ultrasonic signal collides with (hits) an obstacle, the ultrasonic sensor receives a signal reflected from the obstacle.
Implementation Method 3
The TOF value (hereinafter referred to as ultrasonic sensor data) acquired by the ultrasonic sensor is used to estimate the position of an obstacle
Data Source
AI summary
An apparatus for generating a map around a moving object through multi-object recognition using a plurality of ultrasonic sensors is disclosed. The apparatus includes: a sensor unit detecting an object around the moving object; and a processor operatively connected to the sensor unit and configured to generate a mask corresponding to an object around the moving object using data acquired by the sensor unit. The processor is configured to generate a fusion mask using a first mask acquired using each ultrasonic sensor and a prediction mask (hereinafter referred to as a second mask) corresponding to the first mask acquired in a previous cycle; and correct a fusion mask having a common region with the first mask from among fusion masks for all ultrasonic sensors of the sensor unit that has a common region with the first mask.