Vehicle Positioning via Object Sequence Detection in Digital Maps
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional GPS systems for vehicle localization are unreliable in environments with high buildings or tunnels, where satellite signals are blocked, and perform poorly in adverse weather conditions.
Innovation Solution
A method and apparatus that utilize sensing devices like video cameras, lidar, and radar to detect sequences of objects in a digital map, allowing for precise vehicle positioning even without satellite signals, using existing sensors and achieving robustness and high accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If GPS satellite signal is used for vehicle localization, then positioning can be achieved under normal conditions, but positioning fails or becomes unreliable in environments with high buildings, tunnels, or during bad weather
Solution Approach 1:
The system segments the positioning task into multiple independent sensing channels (camera, radar, lidar, ultrasound) that can operate autonomously. Each sensor type detects different objects (lane markings, vehicles, pedestrians, infrastructure) and contributes separately to the overall positioning solution, ensuring that failure of one sensor or signal source does not compromise the entire system.
Solution Approach 2:
The sensing device is designed to perform multiple functions: detecting lane markings for lateral positioning, identifying vehicles and pedestrians for contextual awareness, recognizing infrastructure objects for location verification, and measuring distances for speed calculation. This multi-functional approach allows a single sensor system to provide comprehensive positioning data across diverse environments.
2Reliability
If multiple sensing devices (video camera, lidar, radar, ultrasound) are used to ensure robust object sensing, then positioning reliability improves, but device complexity increases
Solution Approach 1:
The patent merges multiple sensing devices (video camera, radar, lidar, ultrasound) into a single integrated sensing device that operates as one cohesive system. The detection device combines data from all sensor types and processes them uniformly through a single object detection algorithm, managing complexity by treating diverse sensors as a unified multi-functional unit rather than separate systems.
Solution Approach 2:
The system uses objects detected in the surround field (lane markings, vehicles, infrastructure) to automatically determine vehicle position and velocity without requiring external reference systems. The sensing device captures data, the detection device identifies objects and their sequences, and the determination device calculates positioning parameters autonomously, making the system self-sufficient and reducing operational complexity.
3Measurement precision
If object sequence detection in digital map is used for precise positioning, then positioning accuracy improves to 5 cm, but computational complexity increases
Solution Approach 1:
The digital map is pre-populated with information about objects (lane markings, infrastructure, vehicles) and their expected sequences along road sections. This preliminary structuring of spatial data allows the detection device to efficiently match detected object sequences against pre-stored map patterns, reducing real-time computational complexity while maintaining high positioning accuracy of 5 cm.
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
Enables reliable and precise vehicle positioning with high availability, especially in tunnels, cities, forests, and parking garages, with accuracy up to 5 cm, and allows for non-slip determination of linear velocity, even in bad weather.
Implementation Method 1
this is accomplished with the aid of a video camera or a lidar sensor
Implementation Method 2
as well as radar and ultrasound
Implementation Method 3
as well as radar and ultrasound
Data Source
AI summary
A method for determining the position of a vehicle, including sensing of multiple objects in a surround field of the vehicle, detecting a sequence of the multiple sensed objects in a digital map and determining a position of the vehicle in the digital map based on a position of the detected sequence in the digital map. An apparatus for determining the position of a vehicle, as well as to a computer program are also described.