Parking Obstacle Classification for Accurate Vehicle Braking
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current parking collision-avoidance assist systems in autonomous vehicles often result in erroneous braking due to inaccurate obstacle estimation, particularly when distinguishing between normal and low obstacles, which can lead to inappropriate braking timing and increased risk of collisions during parking maneuvers.
Innovation Solution
A vehicle system utilizing an ultrasonic sensor and camera to form map information by assigning weights to occupancy information, determining the type of obstacles based on covariance, and controlling braking accordingly to prevent collisions, with the ability to differentiate between normal and low obstacles and adjust braking timing based on the type and proximity of obstacles.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If obstacle detection is performed using ultrasonic sensor and camera with weighted occupancy information, then measurement precision of obstacle position is improved, but device complexity increases
Solution Approach 1:
The system segments the obstacle detection task by dividing the surrounding space into multiple occupancy grids and processing each grid cell independently with weighted probabilities. This allows complex multi-sensor data to be broken down into manageable discrete units that can be processed separately and then integrated.
Solution Approach 2:
The patent introduces an intermediary processing layer that fuses ultrasonic sensor data and camera image data through occupancy grid mapping. This intermediary representation serves as a mediator between raw sensor inputs and obstacle detection decisions, simplifying the integration of heterogeneous sensor data.
2Reliability
If braking control is activated immediately upon obstacle detection, then collision avoidance reliability is improved, but erroneous braking increases due to false obstacle identification
Solution Approach 1:
The system implements feedback by continuously updating occupancy probabilities based on weighted sensor data and comparing detected obstacles against the probabilistic occupancy map. This feedback mechanism allows the system to verify obstacle detections against multiple data sources before triggering braking, reducing false positives while maintaining collision avoidance reliability.
Solution Approach 2:
The patent performs preliminary classification of detected obstacles as either normal obstacles or low obstacles before activating braking control. This preliminary action distinguishes between obstacle types that require different responses, preventing erroneous braking for low obstacles while maintaining protective braking for normal obstacles.
3Speed
If uniform braking response is applied to all detected obstacles, then response time is reduced, but braking appropriateness deteriorates due to inability to distinguish obstacle types
Solution Approach 1:
The system applies local quality by providing different braking responses based on the local characteristics of detected obstacles. Normal obstacles trigger immediate braking control, while low obstacles trigger delayed braking control. This localized differentiation of response quality maintains fast response for critical obstacles while allowing appropriate delay for non-critical obstacles.
Solution Approach 2:
The patent implements dynamic braking control that adapts the braking timing based on the type of obstacle detected. The braking response is not static but dynamically adjusted according to obstacle classification, allowing the system to optimize response speed for each specific situation rather than using a fixed response time.
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 system effectively inhibits erroneous braking by accurately estimating obstacle positions and types, allowing for timely and appropriate braking responses to prevent collisions, thereby enhancing safety during parking operations.
Implementation Method 1
an ultrasonic sensor configured to acquire occupancy information of a surrounding area of the vehicle based on an ultrasonic signal
Data Source
Figure 1
Figure 2
Figure 3
AI summary
The vehicle includes: a sensor part configured to acquire occupancy information of an surrounding area of the vehicle and a speed of the vehicle; a camera configured to acquire a surrounding image of the vehicle; and a controller configured to form map information based on the occupancy information according to movement of the vehicle, determine presence or absence of an obstacle around the vehicle based on the map information and the surrounding image, and control, in response to presence of the obstacle, the vehicle based on the presence of the obstacle and a possibility of collision of the vehicle derived from the speed of the vehicle and the map information.