Wheel Orientation Sensing for Early Future Heading Prediction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous vehicles face challenges in accurately predicting the future trajectory of objects with wheels, such as vehicles or bicycles, due to the difficulty in detecting thin tires and estimating their orientation, which affects safe navigation and control.
Innovation Solution
The method involves receiving sensor data, identifying the area corresponding to a wheel's expected location, estimating the wheel's orientation by fitting sensor data points to a plane, and determining the object's future heading, which can be used to generate notifications for passengers or control the vehicle autonomously.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If sensor data is used to detect objects and their characteristics, then detection and identification accuracy is improved, but the ability to predict future trajectory accurately deteriorates due to difficulty in detecting wheel orientation
Solution Approach 1:
The patent uses sensor data points as intermediary elements to infer wheel orientation. Instead of directly detecting wheel orientation (which is difficult), the system identifies sensor data points within an expected wheel location area and uses their spatial distribution to estimate the wheel's orientation angle, thereby indirectly obtaining trajectory prediction information.
Solution Approach 2:
The patent replaces direct mechanical/optical detection of wheel orientation with a computational geometry approach. By fitting sensor data points to a plane and calculating angular differences between planes, the system substitutes complex direct measurement with mathematical modeling and data processing.
2Reliability
If the system tries to directly detect wheel orientation, then trajectory prediction may be improved, but detection difficulty increases due to thin tires
Solution Approach 1:
The patent extracts only the necessary information for trajectory prediction without attempting to fully detect the wheel structure. It identifies sensor data points within a predefined expected wheel location area and extracts their spatial coordinates to estimate orientation, avoiding the need to detect the entire wheel structure including thin tires.
Solution Approach 2:
The patent changes the detection parameter from direct wheel orientation measurement to indirect estimation through sensor data point distribution analysis. By transforming the problem from measuring physical wheel properties to analyzing spatial patterns of sensor data points, the system overcomes the detection difficulty of thin tires.
3Reliability
If more sensor data points are collected to improve wheel orientation estimation, then trajectory prediction improves, but data processing complexity and time increase
Solution Approach 1:
The patent segments the sensor data processing by first identifying points within a specific expected wheel location area, then separately fitting these points to a plane. This segmentation allows focused processing on relevant data points rather than all sensor data, reducing overall processing complexity while maintaining accuracy.
Solution Approach 2:
The patent uses only the portion of sensor data points that fall within the expected wheel location area, rather than processing all available sensor data. This partial action approach sufficient for wheel orientation estimation without the excessive processing burden of using all sensor data points.
Data Source
AI summary
The technology relates to determining a future heading of an object. In order to do so, sensor data, including information identifying a bounding box representing an object in a vehicle's environment and locations of sensor data points corresponding to the object, may be received. Based on dimensions of the bounding box, an area corresponding to a wheel of the object may be identified. An orientation of the wheel may then be estimated based on the sensor data points having locations within the area. The estimation may then be used to determine a future heading of the object.