Barrier Shape Estimation From Height-Layered Drone Sensor Data

The patent segments the air space into multiple height layers and processes barrier detection data separately for each layer. By dividing the sensing data into height-based segments, the system can efficiently collect and process barrier information at different altitudes without requiring simultaneous sensing across the entire vertical range, thus improving data collection efficiency while maintaining comprehensive barrier detection coverage.

The patent performs preliminary classification and filtering of sensor data by height layers before full barrier analysis. By pre-organizing sensing data according to height segments and identifying potential barrier regions in advance, the system reduces the computational burden and sensing frequency required, thereby improving data collection efficiency while ensuring reliable barrier detection across all heights.

The patent applies different sensing frequencies and processing intensities to different spatial regions based on barrier probability. Areas with higher likelihood of barriers receive more frequent and detailed sensing, while clear areas use reduced sensing frequency. This localized quality adjustment maintains high barrier detection accuracy in critical regions while reducing overall time loss and energy consumption.

The patent implements partial sensing by focusing computational resources and sensing frequency only on regions where barriers are detected or suspected, rather than performing exhaustive sensing across the entire air space. This partial action approach maintains adequate barrier detection accuracy while significantly reducing the time and energy required for comprehensive sensing.