Vehicle Localization Using Sensor Fusion and Kalman Filtering
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous and semi-autonomous vehicles face navigation challenges due to inaccuracies in location data from sensors, which can increase the risk of collisions with other objects.
Innovation Solution
A system that fuses data from vehicle location sensors with external object detection sensors, such as Lidar or cameras, to determine an adjusted vehicle position by correlating 3D volumes with map data and identifying reference points within these volumes, using Kalman filters to improve data accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If GPS location data is used for vehicle navigation, then the vehicle can obtain location information, but the location data contains inaccuracies that make it difficult to navigate the vehicle precisely
Solution Approach 1:
The patent combines GPS location data with object detection data from external sensors to determine vehicle position. By merging multiple data sources, the system compensates for GPS inaccuracies and achieves more precise and reliable location determination for navigation.
Solution Approach 2:
The patent introduces map data as an intermediary to translate raw GPS coordinates into meaningful location information. The map data serves as a reference framework that helps interpret and correct GPS position data, improving overall location accuracy and navigation reliability.
2Measurement precision
If external object detection sensors are used to improve location accuracy, then the vehicle position determination becomes more precise, but the system complexity increases due to data fusion requirements
Solution Approach 1:
The patent makes the vehicle computer perform multiple functions: it processes GPS data, object detection data, and map data, then integrates all this information to determine vehicle position. This multi-functional approach consolidates data fusion operations into a single processing unit, managing system complexity while improving position accuracy.
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
Enhances the accuracy of vehicle localization, reducing the risk of collisions by providing a more precise determination of the vehicle's position and orientation, even in the presence of sensor inaccuracies.
Implementation Method 1
The instructions may further comprise instructions to filter the vehicle position data by applying a first Kalman filter to the vehicle position data, and to filter position data of the identified point by applying a second Kalman filter to the position data of the point.
Data Source
AI summary
A computer includes a processor and a memory. The memory stores instructions executable by the processor to receive, in a vehicle, object data from an external node, and upon identifying a point, in the received object data, that is within a volume defined based on vehicle position data received from a vehicle sensor, to determine an adjusted vehicle position based on the identified point and the vehicle position data.