Vehicle Radar Sensor Target Segmentation for Computational Load Reduction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing radar sensor systems in vehicles struggle to accurately track targets, especially in complex scenarios like expressways and cities, due to limitations in computational power and accuracy requirements for autonomous driving.
Innovation Solution
A method that uses semantic segmentation to categorize targets into stationary and moving targets, and further classifies moving targets into primary and secondary targets based on distance from the vehicle. Primary targets are processed using high-accuracy algorithms like ETO and RFS, while secondary targets are processed using simpler Kalman filtering, optimizing computational resources.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If complex algorithms (ETO/RFS) are used to track all moving targets, then measurement precision is improved, but device complexity and computational load increase beyond the capability of conventional electronic control units
Solution Approach 1:
The patent segments the set of all moving targets into two distinct subsets: primary targets (within threshold distance) and secondary targets (beyond threshold distance). This segmentation allows different tracking algorithms to be applied to different target groups, resolving the contradiction by enabling high-precision complex algorithms to be used only where necessary (for primary targets) while using simpler algorithms for less critical targets (secondary targets), thus maintaining measurement precision for important targets while reducing overall computational complexity.
Solution Approach 2:
The patent applies the principle of local quality by assigning different algorithmic processing qualities to different spatial regions. Primary targets in the near field receive high-quality complex ETO/RFS algorithm processing, while secondary targets in the far field receive standard-quality Kalman filter processing. This local differentiation resolves the contradiction by concentrating computational resources on locally critical targets rather than uniformly applying complex algorithms to all targets.
2Measurement precision
If complex algorithms (ETO/RFS) are used for all moving targets, then measurement precision is improved, but productivity decreases due to insufficient processing speed for real-time requirements
Solution Approach 1:
The patent segments targets based on distance to enable differential processing: primary targets (close proximity) are processed with computationally intensive ETO/RFS algorithms to ensure high precision, while secondary targets (distant) are processed with faster but less complex Kalman filters. This segmentation resolves the contradiction by achieving high measurement precision only for primary targets where it is most critical, while maintaining overall real-time processing productivity through the use of simpler algorithms for secondary targets.
Solution Approach 2:
The patent applies partial action by using complex high-precision algorithms (ETO/RFS) only partially - specifically for primary targets within the threshold distance - rather than applying them excessively to all targets. This partial application of complex algorithms resolves the contradiction by providing sufficient measurement precision for critical near-field targets while maintaining productivity through simpler algorithms for far-field targets.
3Device complexity
If simple algorithms (Kalman filtering) are used for all targets, then device complexity is reduced, but measurement precision deteriorates causing target merging and loss of autonomous driving accuracy
Solution Approach 1:
The patent segments targets into primary and secondary categories based on distance, enabling the system to use simple Kalman filtering for secondary targets while reserving complex ETO/RFS algorithms for primary targets. This segmentation resolves the contradiction by ensuring that measurement precision is maintained for primary targets through complex algorithms, while allowing device complexity to be reduced through simple algorithms for secondary targets.
Solution Approach 2:
The patent applies local quality by differentiating algorithmic processing quality based on spatial location: complex ETO/RFS algorithms are applied locally to primary targets in the near field where high precision is critical for autonomous driving, while simpler Kalman filters are applied to secondary targets in the far field. This local quality differentiation resolves the contradiction by maintaining measurement precision where it matters most while reducing overall device complexity.
4Measurement precision
If all moving targets are processed with high computational expenditure, then measurement precision is improved, but loss of time increases due to inability to complete processing within real-time constraints
Solution Approach 1:
The patent segments the target processing workload into two time-critical components: primary targets processed with complex ETO/RFS algorithms that require more time but are fewer in number, and secondary targets processed with faster Kalman filters that require less time per target. This segmentation resolves the contradiction by ensuring measurement precision for primary targets while minimizing total processing time through efficient handling of secondary targets.
Solution Approach 2:
The patent applies partial action by using high-computational-expenditure complex algorithms only partially for primary targets rather than for all targets. This partial application resolves the contradiction by achieving sufficient measurement precision for critical primary targets within acceptable time limits, while maintaining overall real-time processing capability through faster algorithms for secondary targets.
Data Source
AI summary
A method for operating radar sensors in a vehicle. At the outset, the acquired targets are divided into stationary targets and moving targets. The moving targets are then divided into primary targets, whereof the distances from the vehicle are less than a pre-definable threshold value, and secondary targets, whereof the distances from the vehicle are greater than a threshold value. The primary targets are fed to a first tracking device, which ascertains the states of the primary targets. The secondary targets are fed to a second tracking device, which ascertains the states of the secondary targets. The second tracking device carries out a computationally less powerful ascertainment of the states than the first tracking device.