Road Object Map Placement with Sensor-Fused Pose Graphs
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for detecting and placing road objects on a map often require expensive hardware or costly computing tasks, making them less scalable and less frequent in updating maps.
Innovation Solution
A method using sensor information to determine initial map positions of road objects, construct a pose graph, and optimize it to minimize edge lengths, leveraging low-cost devices and multiple images to refine camera and object poses.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If expensive hardware or costly computing tasks like SFM are used for detecting and placing road objects, then measurement precision and reliability improve, but device complexity and cost increase
Solution Approach 1:
The patent replaces expensive, specialized hardware with inexpensive, widely available mobile devices (smartphones, tablets) equipped with standard sensors (camera, GPS, accelerometer, gyroscope). These low-cost devices capture images and sensor data that are processed to achieve accurate road object placement, making the system scalable and accessible without requiring specialized equipment
Solution Approach 2:
The patent substitutes complex mechanical/computational systems (Structure from Motion algorithms requiring multiple images and heavy computation) with a sensor fusion approach that integrates data from GPS, accelerometer, gyroscope, and camera. This substitution reduces computational complexity while maintaining accuracy by leveraging readily available sensor data alongside image processing
2Measurement precision
If expensive hardware or costly computing tasks are used, then measurement precision improves, but productivity and scalability worsen
Solution Approach 1:
By using inexpensive mobile devices with standard sensors instead of expensive specialized hardware, the patent enables widespread deployment across many users and devices. This scalability allows frequent map updates as more contributors generate data, increasing productivity without sacrificing placement accuracy
Solution Approach 2:
The system enables ordinary users to contribute to map updates using their personal mobile devices during normal activities. Users automatically capture images and sensor data that are processed to update road object positions, transforming passive map consumers into active contributors and increasing update frequency without requiring dedicated surveyors or expensive equipment
3Measurement precision
If multiple images from different camera positions are used to determine road object positions, then measurement precision improves through error aggregation, but device complexity and processing requirements increase
Solution Approach 1:
The patent replaces complex Structure from Motion processing (requiring multiple images, feature matching, and heavy computation) with a sensor fusion approach. By integrating data from GPS, accelerometer, gyroscope, and camera, the system determines device pose and road object positions with reduced computational requirements while maintaining or improving accuracy through error aggregation from multiple measurements
4Device complexity
If standard mobile phones with cheap hardware are used, then device complexity and cost decrease, but measurement precision and reliability worsen
Solution Approach 1:
The patent combines data from multiple sensors (GPS, accelerometer, gyroscope, camera) and multiple image sources to compensate for the limitations of individual low-cost components. By fusing these diverse data streams and aggregating measurements from multiple images and device positions, the system achieves reliable and accurate road object placement using only inexpensive, widely available mobile devices
Data Source
AI summary
Aspects concern a method including determining initial map positions of a road object on a map, the initial map positions respectively corresponding to images of the road object that are captured by a camera at different camera positions on the map, and constructing a pose graph in which one or more pairs of the initial map positions are respectively connected by first edges, one or more pairs of the different camera positions are respectively connected by second edges, and the initial map positions are respectively connected to the different camera positions by third edges. The method further includes optimizing the constructed pose graph by adjusting the initial map positions, the different camera positions, the second edges and the third edges so that lengths of the first edges are minimized, and determining a final map position of the road object, based on the optimized pose graph.