Adaptive UAV Obstacle Sensing With Weighted Sensor Fusion
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current flight control systems for UAVs are inadequate in detecting and avoiding obstacles, especially non-cooperative targets, due to reliance on transponders and costly radar arrays, which limits their effectiveness in various environments and altitudes.
Innovation Solution
An adaptive sense and avoid system that combines multiple sensors, including echolocation, vision-based, and radar systems, dynamically adjusts sensor modes and weights based on altitude, geographic area, and airspace conditions to provide a 360-degree field of view and optimize obstacle detection and avoidance maneuvers.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If TCAS with transponders is used for obstacle detection, then cooperative aircraft can locate and avoid each other, but non-cooperative obstacles without transponders cannot be detected
Solution Approach 1:
The system employs multiple sensor types (radar, acoustic, visual, GPS/ADS-B) that can detect both cooperative targets with transponders and non-cooperative targets without transponders. Each sensor type serves multiple detection functions, making the overall system universal in its ability to detect various obstacle types through different physical principles and signal methods.
Solution Approach 2:
The patent combines data from multiple independent sensor systems (radar, acoustic sensors, visual sensors, GPS/ADS-B) into a unified obstacle detection and avoidance system. This fusion of multiple detection methods allows the system to overcome the limitations of individual sensors and achieve reliable detection of both cooperative and non-cooperative obstacles.
2Reliability
If radar arrays are used to detect non-cooperative obstacles, then detection capability is improved, but system cost increases significantly
Solution Approach 1:
The system merges radar detection with lower-cost alternative sensors including acoustic sensors, visual sensors, and GPS/ADS-B. This combination allows the system to achieve reliable non-cooperative obstacle detection through multiple methods rather than relying solely on expensive radar arrays, thereby reducing overall system cost while maintaining detection reliability.
Solution Approach 2:
The patent incorporates cheaper sensor alternatives (acoustic sensors, visual sensors, GPS receivers) that can replace or supplement expensive radar systems for certain detection tasks. These lower-cost sensors provide sufficient detection capability for specific scenarios, reducing the need for costly radar infrastructure while maintaining adequate obstacle detection performance.
3Ease of operation
If GPS and obstacle databases are used for obstacle avoidance, then navigation is supported, but GPS accuracy varies widely and many obstructions are not included in databases
Solution Approach 1:
The system combines GPS location data with active sensor detection (radar, acoustic, visual sensors) to determine obstacle positions. Rather than relying solely on GPS and pre-stored databases, the system fuses GPS information with real-time sensor measurements, thereby compensating for GPS accuracy limitations and database incompleteness through direct environmental sensing.
Solution Approach 2:
The system uses real-time sensor feedback to continuously update obstacle location information rather than relying solely on pre-stored database information. The sensors provide continuous feedback about actual environmental obstacles, allowing the system to adapt to unknown or moving obstacles not included in databases and to correct GPS position errors through sensor-based position verification.
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 ability of UAVs to detect and avoid both cooperative and non-cooperative obstacles in diverse environments, minimizing false alarms and ensuring safety by optimizing sensor fusion and trajectory planning based on real-time data and environmental factors.
Implementation Method 1
Attempts have been made, and are ongoing, to utilize small radars, camera-based visual systems, and other detection sensors for non-cooperative targets
Implementation Method 2
current flight control systems designed to detect and avoid non-cooperative obstructions utilize costly radar arrays to track obstacle obstructions
Data Source
Figure 1a~1b
Figure 1c
Figure 2a
AI summary
An adaptive sense and avoid system for use with an aerial vehicle in an environment, the adaptive sense and avoid system comprising a processor, a plurality of sensors, an obstacle detection circuit, and an avoidance trajectory circuit. The processor may be operatively coupled with a flight controller and a memory device, wherein the memory device comprises one or more databases. The plurality of sensors generates sensor data reflecting a position of an obstacle in the environment. The obstacle detection circuit, which is operatively coupled to the processor and the plurality of sensors, may be configured to identify obstacles based at least in part on the sensor data and to generate obstacle information that reflects a best estimate of a position of the obstacle in the environment. The obstacle detection circuit is configured to weigh the sensor data as a function of the aerial vehicle's state and its environment. The avoidance trajectory circuit, which is operatively coupled to the obstacle detection circuit and the processor, may be configured to calculate trajectory data as a function of the obstacle information and information from the one or more databases. The obstacle detection circuit is configured to communicate the trajectory data to the flight controller.