Datacenter UAV Navigation Using Visual Tags After Connectivity Loss

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Monitoring and reporting the operating state of a datacenter in real-time is challenging due to its complex and dynamic nature, especially in environments where GPS signals are weak or absent, and conventional recovery operations for UAVs are inefficient when connectivity is lost.

Innovation Solution

An unmanned aerial vehicle (UAV) executes an autonomous navigation process using visual tags positioned within the datacenter, allowing it to collect data without GPS signals or a constant connection to a server, and switches to a connectivity-enabled navigation process once connectivity is restored, adjusting the data collection path as needed.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Loss of information

If a connectivity-enabled navigation process is used for data collection, then real-time monitoring and reporting capability is improved, but the system becomes vulnerable to connectivity losses and cannot operate in GPS-denied environments

Engineering Contradiction:
Improvereal-time monitoring capabilityVSAvoidoperational continuity during connectivity loss
Core Design Contradiction:
Loss of informationVSReliability

Solution Approach 1:

The navigation system dynamically switches between two operational modes: connectivity-enabled navigation when server communication is available, and autonomous navigation when connectivity is lost. This dynamic adaptation allows the system to maintain operational reliability across varying connectivity conditions while maximizing real-time monitoring capabilities when possible.

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The system changes its operational parameters based on connectivity status. When connected, it operates in real-time monitoring mode with server coordination; when disconnected, it transitions to autonomous mode using pre-planned paths and onboard navigation, effectively changing the operational parameters to match environmental conditions.

Inventive Principle:
Principle #35Parameter changes

2Measurement precision

If GPS-based navigation is used for UAV navigation, then navigation accuracy is improved, but the system becomes inoperative in datacenter environments where GPS signals are weak or absent

Engineering Contradiction:
Improvenavigation accuracyVSAvoidoperability in GPS-denied environments
Core Design Contradiction:
Measurement precisionVSAdaptability or versatility

Solution Approach 1:

Visual tags serve as intermediary reference points within the datacenter environment. These tags are positioned at known locations and provide the UAV with location reference information similar to GPS, enabling accurate navigation and positioning in GPS-denied indoor environments through computer vision-based tag recognition and localization.

Inventive Principle:
Principle #24Intermediary (Mediator)

Solution Approach 2:

The system replaces GPS satellite-based electromagnetic positioning with an onboard computer vision system that detects and processes visual tags. This substitution enables the UAV to achieve navigation accuracy comparable to GPS using visual recognition and image processing technologies instead of satellite signals.

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

3Reliability

If conventional recovery operations are used when connectivity is lost, then the UAV returns to a safe state, but data collection efficiency is reduced due to inefficient recovery paths

Engineering Contradiction:
Improvesafe state recoveryVSAvoiddata collection efficiency
Core Design Contradiction:
ReliabilityVSProductivity

Solution Approach 1:

The system performs preliminary actions by pre-planning navigation paths and identifying visual tags along optimal data collection routes before connectivity is lost. When disconnected, the UAV can continue following these pre-planned paths using onboard autonomous navigation, avoiding the need to return to origin points or docking stations and thereby maintaining data collection efficiency while ensuring safe operation.

Inventive Principle:
Principle #10Preliminary action

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

Enables efficient and resource-effective data collection in datacenters by allowing UAVs to navigate and collect data without GPS, recover from connectivity losses, and reconcile data collection paths, improving datacenter monitoring and reducing resource intensity.

Implementation Method 1

The UAVs are adapted to use said tags as reference positions in the case that GPS signals are weak or absent

Methodology Applied
Scientific EffectOptical recognition: Photography

Data Source

PatentEP4276561B1Autonomous aerial imaging and environmental sensing of a datacenter
Publication Date: 2025.01.01 GOOGLE LLC
  • EP4276561B1 patent drawingFigure 1
  • EP4276561B1 patent drawingFigure 2
  • EP4276561B1 patent drawingFigure 3A~3B

AI summary

Methods, systems and apparatus, including computer programs encoded on computer storage medium, for autonomous aerial imaging and environmental sensing of a datacenter (100, 104a, 104b). In one aspect a method includes executing, within a datacenter and by an unmanned aerial vehicle (102), UAV, in data communication with a server (112), a connectivity-enabled navigation process to collect data (108) from the datacenter; determining, during execution of the connectivity-enabled navigation process and by the UAV, that connection to the server (112) is lost so that the connectivity-enabled navigation process cannot be continued; and in response to determining that connection to the server is lost, executing, by the UAV, an autonomous navigation process until the connection to the server is recovered, wherein the autonomous navigation process navigates the datacenter using visual tags (114) positioned in the datacenter.