Neural Pose Graph Aggregation for SLAM Error Correction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Simultaneous location and mapping (SLAM) algorithms suffer from accumulated errors in pose estimation, leading to increasing inaccuracy over time, which affects the precision of object localization and navigation.
Innovation Solution
A method that uses neural pose graphs to correct accumulated errors by aggregating pose estimations across temporally adjacent time periods, employing an attention phase and sequence encoder operations to synthesize a corrected pose estimation, thereby improving localization accuracy and reducing power consumption.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If SLAM algorithms are used for pose estimation, then localization and mapping can be performed, but accumulated errors grow over time resulting in increasing inaccuracy
Solution Approach 1:
The patent implements feedback by using corrected pose estimations from temporal aggregation to adjust and correct future pose estimations. The system continuously monitors accumulated errors and feeds correction signals back into the pose estimation process, preventing error accumulation and maintaining long-term accuracy in localization and mapping operations.
Solution Approach 2:
The patent applies preliminary action by performing temporal aggregation of pose estimations before final localization decisions are made. By pre-correcting pose estimates using historical data and neural pose graphs, the system proactively prevents error accumulation rather than reacting to it after it occurs, thereby maintaining reliability over extended periods.
2Measurement precision
If neural pose graphs and temporal aggregation are used to correct pose estimation errors, then localization accuracy is improved, but computational complexity and processing requirements increase
Solution Approach 1:
The patent segments the pose estimation process into distinct components: initial pose estimation, neural pose graph construction, temporal aggregation, and correction application. This segmentation allows each component to be optimized independently and enables parallel processing of multiple time steps, reducing overall computational complexity while maintaining high localization accuracy through specialized processing for each function.
Solution Approach 2:
The patent applies partial action by selectively aggregating pose estimations from a limited temporal window rather than processing all historical data. The system uses a fixed number of recent time steps for aggregation, providing sufficient correction without the excessive computational burden of processing entire historical sequences, thus balancing accuracy improvement with manageable complexity.
Data Source
AI summary
In accordance with some embodiments, a method is performed at a device with one or more processors and non-transitory memory. The method includes obtaining location vector data characterizing an object. The method includes determining a neural pose graph associated with a respective time-period based on an initial local pose estimation as a function of respective location vector data. The method includes determining a meta pose estimation associated with the respective time-period by aggregating the neural pose graph associated with the respective time-period and one or more other neural pose graphs associated with one or more temporally adjacent time-periods. The method includes synthesizing a corrected pose estimation by correcting the meta pose estimation associated with the respective time-period based on a function of the meta pose estimation associated with the respective time-period and meta pose estimations associated with one or more temporally adjacent time-periods in order to correct accumulated errors in the initial local pose estimation.