Object Density Classification via Air-Displacement Motion Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Automated vehicles face challenges in distinguishing between objects that require evasive maneuvers and those that do not, leading to unnecessary actions when encountering objects like empty paper bags or small tumbleweeds.
Innovation Solution
An object-classification system that uses an object-detector and controller to determine the density of objects based on motion characteristics caused by air-movement, distinguishing between dense and low-density objects by analyzing steering corrections, other-object movements, and air-displacement models from passing vehicles.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If automated vehicles are programmed to avoid running-over or striking any detected object, then safety is improved, but unnecessary evasive maneuvers occur when encountering low-density objects like empty paper bags or small tumbleweeds
Solution Approach 1:
The system changes the parameter of object classification from binary (object present/absent) to multi-dimensional (density classification based on motion characteristics). By analyzing motion parameters such as displacement, velocity, and acceleration of detected objects, the system categorizes objects into dense and low-density groups, enabling differentiated response strategies that improve operational efficiency while maintaining safety
Solution Approach 2:
The system replaces purely mechanical evasive maneuver execution with an intelligent decision-making process. Instead of automatically executing avoidance maneuvers for all detected objects, the controller uses image processing and motion analysis to substitute mechanical action with cognitive assessment, determining whether evasive maneuvers are actually necessary based on object density classification
2Reliability
If the vehicle performs evasive maneuvers for all detected objects, then collision avoidance is improved, but unnecessary steering corrections and maneuvers are executed
Solution Approach 1:
The system applies different quality standards and response requirements to different regions of the object space. Dense objects trigger full evasive maneuvers while low-density objects trigger no action or minimal adjustments. This local differentiation in response quality allows the vehicle to maintain high collision avoidance for critical objects while avoiding unnecessary maneuvers for insignificant objects
3Measurement precision
If the system classifies objects by density using motion characteristics, then object classification accuracy is improved, but system complexity increases
Solution Approach 1:
The system uses the existing object detector and image processing capabilities for multiple functions: initial object detection, motion characteristic analysis, and density classification. By making the image processing system multi-functional, the patent achieves accurate object classification without adding separate dedicated hardware systems, thereby limiting the increase in overall 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
Enables the automated vehicle to accurately assess the risk of striking objects, avoiding unnecessary maneuvers by differentiating between objects that require avoidance and those that can be safely passed over, thereby improving operational efficiency and safety.
Implementation Method 1
determine a density of the object based on a motion-characteristic of the object caused by air-movement proximate to the object
Data Source
Figure 1
Figure 2
AI summary
An object-classification system (10) for an automated vehicle includes an object-detector (20) and a controller (28). The object-detector (20) may be a camera, radar, lidar or any combination thereof. The object-detector (20) detects an object (18) proximate to a host-vehicle (12). The controller (28) is in communication with the object-detector (20). The controller (28) is configured to determine a density (30) of the object (18) based on a motion-characteristic (32) of the object (18) caused by air-movement (34) proximate to the object (18), and operate the host-vehicle (12) to avoid striking the object (18) with the host-vehicle (12) when the density (30) of the object (18) is classified as dense.