Variable-density scanning strategy and ROI density analysis for high-precision MEMS LiDAR.

Area of Science:

• Optoelectronics and Photonics
• Robotics and Automation
• Micro-Electro-Mechanical Systems (MEMS)

Background:

• Conventional Lissajous scanning in MEMS LiDAR for industrial robotics results in fixed-density point clouds.
• This fixed density leads to sparse resolution in regions of interest (ROI) and redundant data in non-critical areas.
• Addressing these limitations is crucial for improving LiDAR system efficiency and accuracy in industrial applications.

Purpose of the Study:

• To propose a phase modulation (PM) strategy for dynamic adaptive variable-density ROI scanning in MEMS LiDAR.
• To enable localized density enhancement without global redundancy using a single resonant mirror.
• To improve LiDAR system integration and accuracy for industrial robotics.

Main Methods:

• Implementation of a phase modulation (PM) strategy for dynamic adaptive variable-density scanning.
• Utilizing three types of variable density modulation for ROI switching within a nine-grid region.
• Employing kernel density estimation to quantify scanning density and a minimum angular separation method for assessment.

Main Results:

• Achieved a peak scanning density of 10.3596 in ROIs.
• Demonstrated a decrease in angular separation by up to 68.26% and an increase in scanning density by up to 75.52% compared to conventional methods.
• Experimental results confirmed seamless, interference-free transitions between ROI modes, aligning with theoretical predictions.

Conclusions:

• The proposed PM strategy effectively enables dynamic adaptive variable-density scanning in MEMS LiDAR.
• The method enhances localized density in ROIs without global redundancy and avoids the need for multiple mirrors.
• This contributes significantly to improved LiDAR system integration and accuracy in industrial robotics.