Carrier Localization via IMU-Vision Sensor Fusion
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Indoor positioning systems face challenges such as signal shielding in indoor environments and high costs due to the need for external sensors, and they struggle with changing environments and slower positioning speeds compared to external systems.
Innovation Solution
A system and method utilizing both an inertial measurement unit (IMU) and a vision measurement device, such as a mono or stereo camera, to synchronize localization, posture estimation, and environment map establishment in a three-dimensional space, allowing for real-time map building even in dynamic conditions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If an external positioning system is used, then positioning speed is high, but device complexity and cost increase due to requiring external sensors installed beforehand
Solution Approach 1:
The patent combines inertial measurement (IMU) and vision measurement devices into a single integrated system mounted on the carrier. This merging allows the system to achieve both high positioning speed through IMU data and accurate environment mapping through vision data, while eliminating the need for separate external sensors distributed throughout the space.
Solution Approach 2:
The measurement devices mounted on the carrier serve multiple functions simultaneously: they perform carrier localization, posture estimation, environment feature detection, and map establishment. This multi-functionality replaces the need for dedicated external positioning sensors, reducing overall system complexity while maintaining high positioning speed.
2Adaptability or versatility
If an internal positioning system is used, then extensibility is large, but positioning speed is lower and map establishment is required beforehand
Solution Approach 1:
The system performs preliminary environment mapping by using the vision measurement device to detect and record environment features as the carrier moves through the space. This pre-established feature map is then used for subsequent rapid positioning operations, enabling both high extensibility to new environments and fast positioning speed through efficient feature matching.
Solution Approach 2:
The system continuously updates the environment map and carrier position information as movement occurs, rather than requiring complete map establishment beforehand. This continuous operation allows the system to adapt to new environments in real-time while maintaining high positioning speed through ongoing feature detection and tracking.
3Speed
If only inertial measurement device is used, then positioning speed is high, but measurement precision deteriorates due to error accumulation
Solution Approach 1:
The system uses vision measurement results as feedback to correct and verify the position and posture information derived from inertial measurement. The controller compares the carrier's estimated position from IMU data with the position calculated from visual feature matching, and uses this feedback to correct accumulated errors, thereby maintaining both high positioning speed and accurate measurement precision.
Solution Approach 2:
The patent creates a composite positioning solution by fusing data from two different measurement technologies: inertial measurement and vision measurement. This composite approach combines the high-speed advantages of IMU with the high-precision, error-free characteristics of vision-based positioning, achieving both speed and accuracy simultaneously through sensor fusion.
4Measurement precision
If only vision measurement device is used, then measurement precision is high, but positioning speed is lower due to calculation complexity
Solution Approach 1:
The system performs preliminary detection of environment features and establishes their spatial relationships in advance using the vision measurement device. This pre-processing creates a ready-to-use feature map that enables rapid positioning through simple feature matching, rather than requiring complex real-time calculations, thus achieving both high precision and fast positioning speed.
Solution Approach 2:
The system uses inertial measurement to provide partial positioning information that narrows down the search area for vision-based feature matching. This partial action from the IMU reduces the computational burden on the vision system, allowing it to maintain high measurement precision while achieving faster positioning speeds by focusing calculations on a smaller search space.
Data Source
AI summary
Provided is a system for localizing a carrier, estimating a posture of the carrier and establishing a map. The system includes: an inertial measurement device, measuring a motion state and a rotation state of the carrier; a vision measurement device disposed on the carrier for picturing an environment feature in an indoor environment where the carrier locates; and a controller receiving measuring results from the inertial measurement device and the vision measurement device to estimate a posture information, a location information and a velocity information of the carrier, and establishing a map having the environment feature. The controller estimates based on a corrected measuring result from one of the inertial measurement device and the vision measurement device, then controls the other one of the inertial measurement device and vision measurement device to measure, and accordingly corrects the posture, location and velocity information of the carrier and the map.