HAPS Neural Network Navigation for GPS-Drifted Drone Flights
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional drones lack the ability to adapt to unplanned changes in flight paths due to GPS drift and environmental anomalies, leading to potential deviations from the intended course and inefficient data collection.
Innovation Solution
Implementing a neural network on a high-altitude pseudosatellite platform (HAPSNN) to monitor air traffic and assist drones with real-time navigation and obstacle avoidance, and a drone-specific neural network (DINN) to process sensor data and adjust flight paths accordingly.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If conventional drones follow preprogrammed flight paths using GPS waypoints, then navigation is simple and automated, but the drone cannot adapt to GPS drift or unexpected obstacles, leading to course deviations and missed targets
Solution Approach 1:
The patent introduces a HAPS platform as an intermediary between ground control and drones. The HAPS carries onboard neural networks that process sensor data and generate navigation adjustments, which are then transmitted to drones. This intermediary approach enables complex adaptive processing without burdening the drones themselves, resolving the contradiction between adaptability and device complexity.
Solution Approach 2:
The patent replaces conventional mechanical/GPS-based navigation systems with neural network-based cognitive systems. The neural networks on the HAPS platform analyze sensor data and generate adaptive navigation commands, substituting rigid preprogrammed paths with flexible intelligent decision-making, thereby enabling adaptation to GPS drift and environmental changes.
2Productivity
If drones use onboard sensors to monitor the environment, then they can detect obstacles and anomalies, but they lack the capability to process this data in real-time for adaptive navigation and collision avoidance
Solution Approach 1:
The HAPS platform serves as a computational intermediary that receives sensor data from drones, processes it through onboard neural networks, and returns navigation adjustments. This architecture enables real-time adaptive navigation without requiring heavy computational resources on the drones themselves, thus improving productivity while managing device complexity.
Solution Approach 2:
The patent moves the computational processing from the drone dimension to the HAPS dimension. By relocating the neural network processing to the higher-altitude HAPS platform, the system gains real-time adaptive navigation capability while keeping individual drones simpler, effectively using a different operational dimension to resolve the complexity constraint.
3Reliability
If drones operate in authorized airspace under FAA regulations, then flight safety is maintained, but authorization processes are slow and require real-time communication with UTM/LAANC systems
Solution Approach 1:
The HAPS platform pre-calculates and pre-requests airspace authorizations for anticipated drone operations. By performing authorization requests in advance and caching approved corridors, the system maintains reliable airspace operation while eliminating real-time communication delays, thus resolving the contradiction between reliability and time loss.
4Adaptability or versatility
If HAPS platforms provide broad satellite-like coverage, then they can monitor large airspace areas, but they lack the capability to help drones recover from path deviations and prevent data overcollection
Solution Approach 1:
The HAPS platform uses onboard neural networks to continuously monitor drone positions and sensor data, providing feedback control for path recovery. The neural networks analyze deviations from intended paths and generate corrective navigation commands, enabling the HAPS to actively help drones recover from GPS drift and prevent data overcollection, thus enhancing mission-level support capability.
Data Source
AI summary
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.