UAV Neural Navigation for Adaptive Inspection Flight Paths

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Conventional drones lack the ability to autonomously adapt to unplanned changes in their flight path, such as obstacles or GPS drift, which can lead to data overcollection or failure to collect necessary measurements during asset inspection, and require human operators for safe navigation, especially in complex environments like power stations.

Innovation Solution

The implementation of neural networks, both in high-altitude pseudosatellite platforms (HAPSNN) and onboard drones (DINN), which enable real-time data processing and classification, allowing for adaptive flight paths and safe navigation by predicting and responding to changes in the environment, including obstacles and weather conditions, and facilitating communication with air traffic management systems.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If conventional drones follow preprogrammed flight paths with GPS waypoints, then navigation is simple and equipment is basic, but the drone cannot adapt to unplanned changes such as obstacles or GPS drift, leading to safety issues and data collection problems

Engineering Contradiction:
Improveadaptability to unplanned changesVSAvoidsystem complexity
Core Design Contradiction:
Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The drone is equipped with onboard neural networks that enable it to autonomously detect obstacles, correct GPS drift, and adjust flight paths without human intervention. The system processes sensor data and executes navigation decisions independently, making the drone self-sufficient in adapting to unexpected conditions during flight

Inventive Principle:
Principle #25Self-service

Solution Approach 2:

Traditional mechanical navigation systems relying on preprogrammed GPS waypoints are replaced with intelligent software-based neural networks. These neural networks process sensor inputs and generate adaptive flight commands, substituting rigid mechanical control with flexible computational intelligence that can respond to real-time environmental changes

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

2Productivity

If human operators control drones within line of sight for safe navigation, then safety is maintained, but operational efficiency decreases and personnel must be present at each inspection site

Engineering Contradiction:
Improveoperational efficiencyVSAvoidsafety
Core Design Contradiction:
ProductivityVSReliability

Solution Approach 1:

The drone performs self-monitoring and self-correction of its flight path using onboard sensors and neural networks. It automatically detects deviations from the mission path, identifies obstacles, and executes corrective maneuvers without requiring continuous human supervision, thereby maintaining safety while enabling autonomous operation beyond line of sight

Inventive Principle:
Principle #25Self-service

Solution Approach 2:

The drone continuously monitors its position using multiple sensors including GPS, inertial measurement units, and obstacle detection sensors. This real-time feedback is processed by neural networks that compare actual position with planned trajectory, generating corrective commands to maintain safe and efficient flight paths without constant human intervention

Inventive Principle:
Principle #23Feedback

3Measurement precision

If drones closely approach assets for detailed inspection, then data quality improves, but safety risks increase from potential collisions with obstacles or structures

Engineering Contradiction:
Improvedata qualityVSAvoidcollision risk
Core Design Contradiction:
Measurement precisionVSObject-affected harmful factors

Solution Approach 1:

Before the drone approaches the asset for detailed inspection, the neural networks process data from obstacle detection sensors to identify and map potential hazards in the vicinity. This preliminary safety assessment allows the drone to plan a safe approach trajectory that maintains close proximity to the asset for high-quality data collection while avoiding detected obstacles

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The drone's flight path is dynamically adjusted in real-time based on sensor feedback and neural network processing. The system continuously modifies the trajectory to optimize the balance between getting close enough to the asset for high-quality inspection data and maintaining sufficient distance from obstacles to ensure safety, allowing flexible adaptation to changing environmental conditions

Inventive Principle:
Principle #15Dynamics

Data Source

PatentEP4461644A1Neural networks for unmanned aerial vehicles and airborne traffic management
Publication Date: 2024.11.13 DROBOTICS LLC
  • EP4461644A1 patent drawingFigure 1
  • EP4461644A1 patent drawingFigure 2
  • EP4461644A1 patent drawingFigure 3

AI summary

An unmanned aerial vehicle (UAV) or "drone" executes a neural network to assist with inspection, surveillance, reporting, and other missions. The drone inspection neural network may monitor, in real time, the data stream from a plurality of onboard sensors during navigation to an asset along a preprogrammed flight path and/or during its mission (e.g., as it scans and inspects an asset). A HAPS platform may execute a neural network (a "HAPSNN") as it monitors air traffic; the neural network enables it to classify, predict, and resolve events in its airspace of coverage in real time as well as learn from new events that have never before been seen or detected. The HAPSNN-equipped HAPS platform may provide surveillance of nearly 100% of air traffic in its airspace of coverage, and the HAPSNN may process data received from a drone to facilitate safe and efficient drone operation within an airspace.