SLAM Position Estimation via Global-Local Coordinate Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional SLAM-type approaches for estimating a device's position and target positions in complex environments face computational complexity and inconsistency issues, limiting their feasibility and control performance in complex environments.
Innovation Solution
A method that defines a global coordinate reference frame, an estimated local coordinate reference frame, and a correction-free local coordinate reference frame, where the device's position is assumed error-free, allowing for the computation of a state vector that is robust to global corrections and maintains consistency with those corrections.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If SLAM algorithm is used to estimate device position and target positions, then position estimation is achieved, but computational complexity increases as square of number of landmarks
Solution Approach 1:
The patent segments the position estimation problem into two independent parts: (1) estimating the device position using global information from external systems, and (2) estimating target positions using local information relative to the device. This segmentation avoids the computational complexity of SLAM by eliminating the need to simultaneously estimate all positions in a unified coordinate system, thereby resolving the contradiction between estimation accuracy and computational complexity.
2Measurement precision
If SLAM algorithm is used directly in guidance and control loop, then location estimation is provided, but control loop update rate is limited by SLAM update rate
Solution Approach 1:
The patent separates location estimation from the control loop by providing device position information through external global systems rather than through SLAM. This allows the control loop to operate at its full update rate without being constrained by SLAM computational requirements, while still receiving accurate position data for control decisions.
3Adaptability or versatility
If multiple coordinate systems are handled for mapping and manoeuvre planning, then comprehensive navigation information is obtained, but system complexity increases
Solution Approach 1:
The patent extracts the device position estimation function from the local coordinate system processing and handles it separately through global external systems. This extraction eliminates the need to manage multiple coordinate systems simultaneously, as the device position is obtained independently and targets are processed in local coordinates relative to the device, thereby reducing system complexity while maintaining navigation versatility.
Data Source
Figure 1~2B
Figure 3
Figure 4
AI summary
A method of estimating a position of a vehicle (100) and at least one target (204A, 204B) in an environment, the method including computing a state vector defined by: a vector representing a position of the vehicle with respect to an origin of a global coordinate reference frame (g); a vector representing an estimated position of the at least one target with respect to an origin of an estimated local coordinate reference frame (1), and an error state form based on a difference between the origin of the estimated local coordinate reference frame and an origin of a correction-free local coordinate reference frame (L) in which a position of the vehicle is assumed to be error-free. The method comprises implementing a SLAM-type algorithm and receiving a GPS locating signal.