Lidar Spatio-Temporal Filtering for False Return Suppression
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Lidar systems face challenges in distinguishing real object returns from false alarms, which clutter the image, distort object outlines, and can lead to incorrect decision-making, particularly in autonomous systems, due to the noisy nature of photodetectors.
Innovation Solution
Implement spatio-temporal filtering methods that utilize spatial and temporal information to enhance the probability of a measurement being a real return by considering spatial adjacency and historical data, adjusting probabilities across iterations, and thresholding based on these metrics to reduce false alarms.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the detection threshold is lowered to increase maximum range, then the detection capability is improved, but the false alarm rate increases
Solution Approach 1:
The system performs preliminary spatial adjacency checks before final detection decisions. By pre-evaluating whether measurements have spatially adjacent neighbors and maintaining probability metrics over time, the system prepares contextual information that will be used to filter false alarms before they affect the final detection threshold.
Solution Approach 2:
The patent introduces an intermediary probability metric that mediates between the detection threshold and false alarm filtering. This probability metric, updated based on spatial adjacency and temporal persistence, acts as a filter that allows the detection threshold to remain low for maximum range while preventing false alarms from being accepted.
2Reliability
If spatial adjacency information is processed to reduce false alarms, then the reliability is improved, but the processing complexity increases
Solution Approach 1:
The processing is segmented into distinct operations: (1) identifying spatially adjacent measurements within a defined search area, (2) updating probability metrics based on adjacency findings, and (3) thresholding final detections. This segmentation allows each operation to be optimized independently and reduces overall complexity by breaking down the filtering task into manageable steps.
Solution Approach 2:
The system uses the spatial adjacency information and probability metrics to automatically filter false alarms without requiring external intervention or complex control logic. The probability metric updates itself based on the presence of adjacent returns, and the filtering decision is made automatically when the threshold is reached, reducing the need for additional processing complexity.
3Measurement precision
If multiple measurements are processed to identify spatially adjacent returns, then the measurement precision is improved, but the processing time increases
Solution Approach 1:
The system performs preliminary identification of spatially adjacent measurements and updates probability metrics in advance, during the data processing pipeline. By preparing this contextual information before final detection decisions are made, the system reduces the processing time required for final filtering while maintaining high measurement precision through thorough validation.
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
This approach effectively reduces false alarms by providing context to each measurement, improving the accuracy of lidar systems by ensuring that only probable real returns are displayed, thereby enhancing the reliability of the image data.
Implementation Method 1
Photodetectors convert light into measurable electric current
Implementation Method 2
Time-of-flight laser ranging systems generally work by emitting a laser pulse and recording the time it takes for the laser pulse to travel to a target, reflect, and return to a photoreceiver
Implementation Method 3
A signal on the photodetector originating from transmitted laser light which has reflected off an object is called a return
Data Source
AI summary
Methods and apparatus for a lidar system having spatio-temporal filtering to reduce false alarms in image data. In embodiments, the probability of a lidar return being real and not a false alarm is calculated based on both the current and historical presence of other returns which are spatially adjacent to the return being calculated. The probability is used to filter false alarms through thresholding.