Multispectral Sensor Fusion for Autonomous Flight Positioning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional sensor systems for aircraft, designed for human consumption, face limitations in dynamic range and processing power, which can hinder object identification and navigation, especially in challenging environments like fog, due to constraints on color spectrum and frame rate, leading to potential inability to distinguish critical objects like approach lights.
Innovation Solution
A multispectral sensor suite on autonomous aircraft, comprising a vision system, RF RADAR, and LIDAR, processes data streams with varying frame rates and latencies suitable for flight control and positioning, using a processor and database to identify objects and determine trajectories, enabling precise navigation and control.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If traditional human-oriented sensors are used with limited dynamic range and color spectrum, then the system is easier to operate for human pilots, but the ability to distinguish objects in adverse conditions deteriorates
Solution Approach 1:
The patent segments the sensor system into multiple independent spectral channels (e.g., green, red, blue, and infrared sensors) that operate simultaneously. Each sensor captures data in its optimal spectral range with appropriate dynamic range, allowing the system to process multiple spectral bands independently before presenting integrated information to the pilot or autonomous system.
Solution Approach 2:
The patent extends the sensor system into the infrared spectral dimension beyond the visible range. By adding infrared sensors that operate independently with their own dynamic range optimized for thermal detection, the system gains an additional dimensional capability to distinguish objects (like approach lights) that are invisible or indistinguishable in the visible spectrum alone.
2Measurement precision
If high-dynamic-range imagery is processed without perspective transformation, then object identification accuracy improves, but processing complexity increases
Solution Approach 1:
The patent extracts and processes each spectral channel independently through dedicated processing pipelines. Rather than attempting to transform and integrate high-dynamic-range data from multiple sensors into a single perspective-corrected image, the system separates the processing tasks: each sensor's data is processed independently to extract relevant features, then these features are integrated for object identification, avoiding the computational burden of full perspective transformation of HDR imagery.
3Measurement precision
If multiple spectral bands are captured simultaneously, then object distinction capability improves, but hardware weight and processing power requirements increase
Solution Approach 1:
The patent merges multiple spectral sensing capabilities into a single integrated sensor suite with common mounting, power, and data processing infrastructure. By combining green, red, blue, and infrared sensors into a unified system with shared mechanical and electrical subsystems, the patent reduces the overall weight penalty compared to separate sensor systems while maintaining the ability to capture multiple spectral bands simultaneously.
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
The system effectively processes high-dynamic-range imagery without perspective transformation, allowing accurate object identification and trajectory determination, even in adverse conditions, enhancing the autonomous aircraft's ability to navigate and control flight paths.
Implementation Method 1
a laser imaging detection and ranging (LIDAR) system
Implementation Method 2
a radio frequency (RF) radio detection and ranging (RADAR) system
Data Source
AI summary
A system and method are disclosed for design of a suite of multispectral (MS) sensors and processing of enhanced data streams produced by the sensors for autonomous aircraft flight. The suite of MS sensors is specifically configured to produce data streams for processing by an autonomous aircraft object identification and positioning system processor. Multiple, diverse MS sensors image naturally occurring, or artificial features (towers buildings etc.) and produce data streams containing details which are routinely processed by the object identification and positioning system yet would be unrecognizable to a human pilot. The object identification and positioning system correlates MS sensor output with a-priori information stored onboard to determine position and trajectory of the autonomous aircraft. Once position and trajectory are known, the object identification and positioning system sends the data to the autonomous aircraft flight management system for autopilot control of the autonomous aircraft.