Scanning LiDAR Velocity Estimation Across Adjacent Sweeps
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing LiDAR systems struggle to effectively determine the motion of detected objects with sufficient accuracy and low latency, which is critical for applications like autonomous driving.
Innovation Solution
A scanning laser device that interpolates distance measurements from temporally adjacent sweeps to estimate radial velocities for multiple measurement points, allowing for high-frequency and low-latency velocity information.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional LiDAR systems use conventional sweep processing methods, then the system complexity remains low, but the velocity estimation accuracy and latency are insufficient
Solution Approach 1:
The patent segments the LiDAR measurement process into multiple sweeps (first sweep, second sweep, third sweep) with different scan directions. Each sweep processes distance measurements separately, and the velocity estimation is performed by comparing results across these segmented sweeps. This segmentation enables accurate velocity estimation through temporal differentiation while maintaining manageable processing complexity through structured approach.
Solution Approach 2:
The patent performs preliminary distance measurement sweeps before final velocity estimation. Multiple preliminary sweeps are conducted in opposite directions to gather sufficient distance data points. These preliminary actions accumulate the necessary measurement data, and only after this data collection phase does the system perform the velocity estimation calculation, ensuring accuracy while managing processing load.
2Productivity
If LiDAR systems increase the frequency of velocity measurements, then the responsiveness for autonomous driving decisions improves, but the processing time and computational load increase
Solution Approach 1:
The patent implements continuous velocity estimation by performing multiple sweeps in rapid succession. The first, second, and third sweeps are executed continuously with minimal interruption, allowing the system to maintain high measurement frequency. The temporal differentiation between these continuous sweeps provides the basis for velocity calculation while minimizing processing latency through overlapping operations.
Solution Approach 2:
The patent employs periodic sweeping patterns where the LiDAR system alternates between different scan directions (first sweep direction, second sweep direction opposite to first, third sweep direction). This periodic action creates a rhythm of measurement collection that enables consistent velocity estimation at high frequency while the periodic nature allows for optimized processing cycles that reduce latency.
3Measurement precision
If the LiDAR system processes distance measurements from multiple temporally adjacent sweeps, then velocity estimation accuracy improves, but the data processing complexity increases
Solution Approach 1:
The patent segments the data processing into distinct phases: first processing distance measurements from the first sweep, then processing measurements from the second sweep, and finally processing the third sweep. Each phase operates on isolated data sets, and the velocity estimation is performed by comparing the segmented results. This segmentation reduces overall processing complexity by breaking down a complex multi-step process into manageable sequential operations.
Solution Approach 2:
The patent uses the relationship between opposite-direction sweeps to derive velocity information. By comparing distance measurements from sweeps in opposite directions (first and second sweeps being opposite to each other), the system can infer velocity through the temporal differentiation of these reversed measurement sequences. This inversion approach provides velocity estimation while managing processing complexity through the systematic comparison of reversed data sets.
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
Enables accurate differentiation between stationary and moving objects, enhancing the safety and efficiency of autonomous systems by providing timely velocity information.
Implementation Method 1
scan measurement points with laser light pulses to generate a first set of distance measurements based on times-of-flight of reflected light
Implementation Method 2
compare the generated distance estimates to distance measurements taken in the other sweep along the slow-scan axis to determine radial velocity estimates
Data Source
AI summary
The embodiments described herein provide systems and methods that can facilitate improved velocity estimation in light detection and ranging (LiDAR) systems and other scanning laser devices. Specifically, the systems and methods utilize laser light pulses to determine estimates of velocity for multiple measurement points in a scanned region. For example, a scanning laser device can be adapted to scan measurement points during temporally adjacent measurement sweeps and generate distance measurements based on the scans made during those sweeps. The scanning laser device is further adapted to interpolate distance measurements to determine distance estimates for measurement points not directly scanned during at least one of the sweeps, and to compare the generated distance estimates to distance measurements taken in the other sweep to determine radial velocity estimates for corresponding measurement points based on the comparison.