SLAM Feature Selection for Latency and Precision Trade-off

Solution Overview

Problem

Simultaneous localization and mapping (SLAM) technologies face challenges in maintaining real-time performance while ensuring the precision of estimated surrounding map and current pose information, often resulting in latency due to increased computational demands as the number of extracted features increases.

Innovation Solution

A SLAM-based electronic device that selectively chooses features based on a calculated registration error score, limiting the number of features to a reference number to balance precision and real-time performance, using a processor to extract features, calculate scores, and prioritize those with significant influence on registration error.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If the number of features extracted from the front-end increases, then precision of the surrounding map information and current pose information is improved, but latency occurs during the estimation process

Engineering Contradiction:

Improveprecision of surrounding map information and current pose informationVSAvoidlatency during estimation

Core Design Contradiction:

Measurement precisionVSLoss of time

Solution Approach 1:

The patent extracts only the most essential and informative features from the extracted feature set, removing redundant features that contribute minimally to precision but significantly to computational load. This selective extraction maintains measurement precision while reducing the number of features processed by the back-end, thereby decreasing latency.

Inventive Principle:

Principle #2Taking out (Extraction)

Solution Approach 2:

The patent applies different quality criteria to different features, prioritizing features with higher information content and lower redundancy. By assigning different weights or selection priorities to different features based on their local quality (information contribution), the system maintains overall precision while reducing total feature count to minimize processing time.

Inventive Principle:

Principle #3Local quality

2Measurement precision

If the number of features extracted from the front-end increases, then precision of the surrounding map information and current pose information is improved, but calculation amount of the back-end increases

Engineering Contradiction:

Improveprecision of surrounding map information and current pose informationVSAvoidcalculation amount of the back-end

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The patent extracts only the most essential and informative features from the extracted feature set, removing redundant features that contribute minimally to precision but significantly to computational load. This selective extraction maintains measurement precision while reducing the number of features processed by the back-end, thereby decreasing latency.

Inventive Principle:

Principle #2Taking out (Extraction)

Solution Approach 2:

The patent applies partial action by processing only a subset of extracted features through the computationally intensive back-end algorithms. By identifying and processing only the most critical features (partial action) rather than all extracted features, the system achieves sufficient precision with reduced calculation amount.

Inventive Principle:

Principle #16Partial or excessive action

3Measurement precision

If the number of features extracted from the front-end increases, then precision of the surrounding map information and current pose information is improved, but real-time performance is deteriorated

Engineering Contradiction:

Improveprecision of surrounding map information and current pose informationVSAvoidreal-time performance

Core Design Contradiction:

Measurement precisionVSProductivity

Solution Approach 1:

The patent extracts only the most essential and informative features from the extracted feature set, removing redundant features that contribute minimally to precision but significantly to computational load. This selective extraction maintains measurement precision while reducing the number of features processed by the back-end, thereby decreasing latency.

Inventive Principle:

Principle #2Taking out (Extraction)

Solution Approach 2:

The patent changes the parameter of feature count from a high value to an optimized value that balances precision and real-time performance. By adjusting this key parameter based on system capabilities and performance requirements, the system achieves optimal real-time performance while maintaining sufficient precision for the application.

Inventive Principle:

Principle #35Parameter changes

Data Source

PatentUS11756270B2Slam-based electronic device and an operating method thereof

Publication Date: 2023.09.12 SAMSUNG ELECTRONICS CO LTD

AI summary

A simultaneous localization and mapping-based electronic device includes: a data acquisition device configured to acquire external data; a memory; and a processor configured to be operatively connected to the data acquisition device and the memory, wherein the processor is further configured to extract features of surrounding objects from the acquired external data, calculate a score of a registration error of the extracted features when the number of the extracted features is greater than a set number stored in the memory, and select the set number of features from the among the extracted features, based on the calculated score.