Virtual Plane SLAM for More Accurate Robotic Navigation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for autonomous navigation of robotic units in SLAM (Simultaneous Localization and Mapping) struggle to accurately determine position and movement within complex environments, as they rely solely on detected virtual points without effectively utilizing geometric structures and planes, leading to inaccuracies in trajectory determination and representation of surroundings.
Innovation Solution
The method involves detecting virtual points and using a control and/or regulation unit to ascertain virtual planes, which are then used to solve optimization problems in graph theory, incorporating projected points as boundary conditions to enhance the accuracy of position and movement determination, thereby improving autonomous navigation by leveraging geometric structures and planes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If virtual points are detected without utilizing geometric structures and planes, then the system complexity remains low, but the position and movement determination accuracy deteriorates
Solution Approach 1:
The system segments the environment into distinct geometric planes (e.g., floor, walls, ceilings) from the detected virtual points. By dividing the complex spatial data into planar segments, the system can process and utilize geometric structures efficiently, improving position and movement determination accuracy without overwhelming system complexity
Solution Approach 2:
The system transitions from processing only point coordinates to incorporating planar geometric dimensions. By ascertaining virtual planes that define spatial boundaries and surfaces, the system adds dimensional context (plane equations, normal vectors, distances) to the navigation data, thereby improving accuracy while maintaining manageable complexity through structured geometric representation
2Measurement precision
If virtual planes are ascertained and used for solving optimization problems, then the trajectory determination accuracy is improved, but the computing power requirement increases
Solution Approach 1:
The system performs preliminary ascertainment of virtual planes from detected virtual points before solving the optimization problem. By pre-processing the spatial data to establish geometric plane structures, the optimization algorithm can utilize these pre-defined constraints and boundaries, improving trajectory accuracy while reducing the computational burden during the actual navigation optimization process
Solution Approach 2:
The system transforms the optimization problem by incorporating plane parameters (equations, normal vectors, distances) as additional constraints or variables. This parameter enrichment allows the optimization to achieve higher trajectory determination accuracy by leveraging geometric information, while the structured parameter representation keeps the computational complexity manageable through efficient mathematical formulations
Data Source
AI summary
A method for autonomous navigation of a movable robotic unit, in particular at least as part of a SLAM method. A plurality of virtual points from the surroundings around the detection unit is detected using a detection unit. An optimization problem of graph theory is solved using a control and/or regulation unit for ascertaining a position and/or a movement of the robotic unit and/or for detecting the surroundings as a function of the detected virtual points. At least one virtual plane is ascertained, using the control and/or regulation unit, as a function of at least one group of virtual points of the plurality of virtual points. For ascertaining a position and/or a movement of the robotic unit and/or for representing the surroundings, the ascertained virtual plane is used for solving the optimization problem of graph theory using the control and/or regulation unit.