Object Track Adjustment Using Sensor Data and ML Parameters
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing systems for autonomous vehicles face challenges in efficiently updating tracking and identifying of dynamic objects due to delays and inefficient processing when collecting and associating new measurement data from the whole scene with respective tracks of objects.
Innovation Solution
A method and system that utilize a machine learning model to generate an adjustment to track data associated with an object by receiving measurement data from sensors and previous track data, allowing for focused update processing without associating new measurement data with the whole scene.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If new measurement data from the whole scene is collected and associated with respective tracks of objects, then tracking accuracy is improved, but processing time and system complexity increase
Solution Approach 1:
The patent extracts and processes only the measurement data relevant to the specific object being tracked, rather than processing all measurement data from the entire scene. This selective extraction approach maintains tracking accuracy by focusing on object-specific data while significantly reducing the overall processing time and computational burden.
Solution Approach 2:
The patent segments the tracking process into distinct components: obtaining previous track data, extracting current measurement data for the specific object, generating parameter adjustments through machine learning, and updating the track. This segmentation allows each component to be optimized independently, improving overall efficiency while maintaining accuracy.
2Measurement precision
If new measurement data from the whole scene is collected and associated with respective tracks of objects, then tracking accuracy is improved, but device complexity increases
Solution Approach 1:
The system extracts only the necessary measurement data for the specific tracked object rather than processing all scene data. This extraction principle simplifies the system architecture by eliminating unnecessary data processing pathways and reducing the complexity of data association algorithms while maintaining tracking accuracy.
Solution Approach 2:
The patent transforms the tracking approach by changing from direct data association to parameter adjustment. Instead of associating measurement data with tracks, the system generates adjustments to track parameters based on measurement data and previous track information. This parameter-based approach simplifies the system complexity while maintaining tracking accuracy.
3Productivity
If focused update processing is performed on specific object tracking state, then processing efficiency is improved, but measurement precision may be reduced
Solution Approach 1:
The patent changes the processing paradigm from direct measurement association to parameter adjustment. The machine learning model generates adjustments to track parameters based on current measurement data and previous track data, maintaining measurement precision through learned parameter relationships while achieving efficient focused processing of specific objects.
Solution Approach 2:
The system incorporates feedback by using previous track data alongside current measurement data to generate parameter adjustments. This feedback mechanism ensures that focused update processing maintains tracking accuracy by continuously refining track parameters based on both historical and current information, rather than relying solely on single-frame measurements.
Data Source
AI summary
A method includes obtaining a first track associated with an object. A first set of parameters is generated based on the first track. Measurement data are obtained from one or more sensors. A first set of features are extracted from the measurement data. Based on the first set of parameters and the first set of features, a second set of parameters are generated by a machine learning model. The second set of parameters represent an adjustment to the first set of parameters. Based on the second set of parameters, the first track is adjusted to generate a second track associated with the object. The second track is provided to an autonomous vehicle control system for autonomous control of a vehicle.