Sensor Fusion Driver Assistance for Low-Speed Obstacle Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional driver assistance systems are ineffective at low speeds, particularly in areas like spiral parking lots and densely populated districts, due to limited field of view and radar reflection characteristics, and struggle with recognizing obstacles and pedestrians, especially in low-light conditions, leading to frequent accidents.
Innovation Solution
A driver assistance apparatus and method using a fusion of camera, radar, and ultrasonic sensors to control steering and braking, determining driving modes based on vehicle speed, pedestrian density, and environment, and estimating driving paths to prevent collisions with proximity obstacles and pedestrians.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If conventional driver assistance systems use radar and camera with limited field of view, then the system complexity is reduced, but the recognition rate of proximity obstacles and pedestrians deteriorates
Solution Approach 1:
The patent combines multiple sensors (ultrasonic sensors, radar, and camera) into a fused sensing system. The ultrasonic sensors detect proximity obstacles in the front and side areas, radar provides mid-range detection, and camera captures visual information. By merging these sensor inputs, the system achieves comprehensive coverage of proximity obstacles and pedestrians that no single sensor could detect alone, resolving the contradiction between system complexity and recognition rate.
Solution Approach 2:
The patent divides the detection task into multiple segments handled by different sensors based on their strengths. Ultrasonic sensors handle close-range detection in front and side zones, radar handles mid-range vehicle detection, and camera handles pedestrian and environmental recognition. This segmentation allows each sensor to operate in its optimal range, improving overall detection accuracy without requiring a single complex sensor system.
2Ease of operation
If conventional systems operate without sensor fusion, then the ease of operation is improved, but the ability to recognize obstacles in low-light and densely populated areas deteriorates
Solution Approach 1:
The system merges data from ultrasonic sensors, radar, and camera through sensor fusion algorithms. In low-light conditions, the camera may have reduced performance, but the ultrasonic and radar sensors continue to provide reliable detection. The fusion combines these complementary data sources, maintaining high recognition reliability across varying lighting conditions while keeping the system easy to operate through automatic sensor selection and fusion.
3Device complexity
If conventional systems use short-range field of view, then the device complexity is reduced, but the ability to detect lateral obstacles and pedestrians deteriorates
Solution Approach 1:
The patent segments the detection field into multiple zones (front, side, and lateral areas) and assigns ultrasonic sensors to cover the front and side proximity zones. This segmentation extends the effective detection area without requiring a single complex wide-angle sensor, maintaining relative simplicity while achieving comprehensive lateral and forward coverage.
Solution Approach 2:
The system adds spatial dimensionality to detection by positioning ultrasonic sensors in multiple locations (front and sides) to create three-dimensional coverage of the proximity zone. This multi-dimensional sensor arrangement detects lateral obstacles and pedestrians that would be invisible to a single forward-facing sensor, expanding the effective field of view without proportionally increasing system complexity.
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 solution effectively reduces the likelihood of accidents by accurately detecting and avoiding obstacles and pedestrians, especially in low-speed and densely populated areas, enhancing safety and reducing social costs through precise control of vehicle steering and braking.
Implementation Method 1
measuring a distance between the vehicle and an obstacle in the estimated driving path
Implementation Method 2
using a camera, a radar, and an ultrasonic sensor fusion
Implementation Method 3
uses a camera, a radar, a LiDAR, and the like to recognize environment and obstacles around the vehicle
Implementation Method 4
do not recognize proximity distance obstacles from the front and sides of a vehicle due to a limitation of short-range field of view (FOV) and radar reflection characteristics
Data Source
AI summary
A method for operating a driver assistance apparatus includes: recognizing a driving environment of a vehicle; determining a driving mode based on the driving environment; determining whether the vehicle is driven using at least one sensor based on the driving mode; and controlling steering and braking of the vehicle based on whether the vehicle is driven using the at least one sensor.