Multi-Sensor Fusion for Intelligent Driving Vehicles
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Intelligent driving vehicles equipped with single sensors struggle to adapt to complex environments due to limitations in perception and cognition, particularly with millimeter-wave and laser radar sensors, which face issues like lateral position deviation, sensitivity to weather conditions, and data association challenges leading to processing delays and combinatorial explosions.
Innovation Solution
A multi-sensor fusion method and system that uses a Gaussian mixture probability hypothesis density (GM-PHD) algorithm for extended target tracking with millimeter-wave radar, a bounding-box detector with interacting multiple model-unscented Kalman filter (IMM-UKF) for laser radar, and information matrix fusion (IMF) for central data association, enabling the integration of two-dimensional and three-dimensional detection information to obtain global track information.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If multiple sensors are used to improve perception and cognition ability, then the adaptability to complex driving environments is improved, but the data association complexity and processing delay increase due to combinatorial explosion
Solution Approach 1:
The patent segments the data association process by introducing an association window that divides the extended target into multiple sub-regions. Each sub-region is independently associated with sensor measurements, transforming a single complex combinatorial problem into multiple simpler sub-problems. This segmentation reduces the computational complexity from exponential to polynomial scale while maintaining tracking accuracy.
Solution Approach 2:
The patent applies preliminary action by pre-establishing an association window around the predicted target position before performing data association. This window pre-defines the search space and filters out irrelevant measurements in advance, reducing the number of candidate associations and preventing combinatorial explosion before the actual association computation begins.
2Measurement precision
If multiple sensors are used to improve perception accuracy, then the measurement precision is improved, but the processing delay increases due to time-space conversion and central fusion
Solution Approach 1:
The patent performs time-space conversion and coordinate transformation as preliminary actions before the main fusion computation. By pre-aligning the coordinate systems and time stamps of different sensors, the patent eliminates the need for repeated transformations during the fusion process, significantly reducing processing delay while maintaining the accuracy benefits of multi-sensor integration.
3Length of stationary object
If millimeter-wave radar is used for detection, then the detection range is improved, but the lateral position deviation and heading-angle deviation increase
Solution Approach 1:
The patent merges the data from millimeter-wave radar and laser radar through central fusion. The millimeter-wave radar provides long-range detection capability while the laser radar provides high-precision lateral position and heading information. By combining these complementary data sources, the system achieves both extended detection range and improved positional accuracy, overcoming the limitations of individual sensors.
4Measurement precision
If laser radar is used for detection, then the lateral position accuracy is improved, but the speed sensitivity decreases due to weather effects
Solution Approach 1:
The patent combines laser radar and millimeter-wave radar data through fusion. The laser radar provides accurate lateral position information when conditions permit, while the millimeter-wave radar provides reliable speed detection that is less affected by weather. The fusion algorithm dynamically weights contributions from each sensor based on their respective strengths and current operational conditions, ensuring reliable speed detection regardless of weather effects on the laser radar.
Data Source
AI summary
A multi-sensor fusion method and system for intelligent driving vehicles are provided, which relate to the technical field of intelligent driving vehicles. The method includes: establishing an extended target tracker based on a GM-PHD algorithm and a rectangular target model of a detected object; processing detection information of a millimeter-wave radar by using the extended target tracker to obtain millimeter-wave radar track information of the detected object; processing detection information of a laser radar by using the established bounding-box detector and a JPDA tracker provided with an IMM-UKF to obtain laser-radar track information of the detected object; processing the millimeter-wave radar track information and the laser-radar track information by performing time-space conversion to obtain a central fusion node; and processing the central fusion node by using an IMF algorithm to obtain global track information.