Indoor Robot Navigation Using Static Feature Weighting

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Indoor mobile industrial robots face challenges in navigating environments due to their inability to effectively recognize and adapt to changes, particularly when static features are removed or become dynamic, leading to inaccuracies in positioning and collision avoidance.

Innovation Solution

A navigation system that employs remote sensors to differentiate between static and dynamic features by assigning weights based on their likelihood of movement, using a combination of technologies like stereo vision, lidar, and radar, and a controller to determine the robot's position solely from static features, creating a map that updates with recognized static features for reliable navigation.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If the robot uses features detected in the environment for navigation, then it can establish position and navigate, but when those features are removed or moved, the robot cannot adequately react or maintain accurate positioning

Engineering Contradiction:
Improvenavigation reliabilityVSAvoidadaptability to dynamic changes
Core Design Contradiction:
ReliabilityVSAdaptability or versatility

Solution Approach 1:

The system dynamically classifies features as static or dynamic based on temporal analysis of sensor data. Features are re-evaluated across multiple time points, and their classification can change as the robot observes their behavior over time. This allows the navigation system to adapt to changing environmental conditions while maintaining reliable positioning using identified static features.

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The system continuously monitors detected features and provides feedback to update their classification status. When a feature's position changes between observations, this feedback triggers a reclassification from static to dynamic, allowing the robot to adjust its navigation strategy accordingly and maintain accurate positioning.

Inventive Principle:
Principle #23Feedback

2Measurement precision

If the robot assumes all detected features are static for SLAM, then positioning can be established, but accuracy deteriorates when dynamic objects are present

Engineering Contradiction:
Improvepositioning accuracyVSAvoidfeature classification complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The system segments features into distinct categories (static and dynamic) based on their temporal behavior. By separating the analysis of static features from dynamic features, the system can maintain high positioning accuracy using only static features for SLAM while avoiding the complexity of continuously analyzing all features for movement.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The system performs preliminary classification of features as static or dynamic before using them for navigation. This advance classification allows the robot to pre-identify reliable positioning features, improving measurement precision without requiring complex real-time analysis during navigation operations.

Inventive Principle:
Principle #10Preliminary action

3Adaptability or versatility

If the robot navigates using dynamic features, then it can adapt to current environment, but positioning reliability decreases as these features may move

Engineering Contradiction:
Improveenvironmental adaptabilityVSAvoidpositioning reliability
Core Design Contradiction:
Adaptability or versatilityVSReliability

Solution Approach 1:

The system introduces a classification mechanism as an intermediary between feature detection and navigation. This intermediary evaluates feature stability and routes static features to the SLAM positioning system while handling dynamic features separately for obstacle avoidance, thus maintaining positioning reliability while preserving environmental adaptability.

Inventive Principle:
Principle #24Intermediary (Mediator)

Applied Scientific Principles

This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.

Function Achieved in This Case

Enhances the robot's ability to accurately determine its position and avoid collisions by prioritizing static features, improving navigation reliability and efficiency in dynamic environments.

Implementation Method 1

one or more remote sensors... capable of sensing features remote to or at a distance from the sensors

Methodology Applied
Scientific EffectLight reflection: Reflection

Implementation Method 2

a combination of technologies like stereo vision, lidar, and radar

Methodology Applied
Scientific EffectLIDAR: LIDAR

Implementation Method 3

Many algorithms for Simultaneous Localization and Mapping (SLAM) initially assume that what is detected is static

Methodology Applied
Scientific EffectSimultaneous Localization and Mapping (SLAM):

Data Source

PatentEP3979032B1Systems and methods for enabling navigation in environments with dynamic objects
Publication Date: 2024.08.14 CARNEGIE ROBOTICS LLC
  • EP3979032B1 patent drawingFigure 1~2
  • EP3979032B1 patent drawingFigure 3
  • EP3979032B1 patent drawingFigure 4A~4B

AI summary

An indoor mobile industrial robot system configured to provide a weight to a detected object within an operating environment, where the weight relates to how static the feature is. The indoor mobile industrial robot system including a mechanism configured to translate reflected light energy and positional information into a set of data points representing the detected object having at least one of Cartesian and/or polar coordinates, and an intensity, wherein if any discrete data point within the set of data points representing the detected object has an intensity at or above a defined threshold the entire set of data points is converted into a weight and potentially classified representing a static feature, otherwise such set of data points is classified as representing a dynamic feature having a lower weight.