3D Printed Polymer Electromagnetic Scanning Micromirror for LiDAR Applications = 라이다 응용을 위한 3D 프린팅 폴리머 기반 전자력 구동 스캐닝 마이크로미러

다국어 초록 (Multilingual Abstract)

This research developed a three-dimensional (3D) printed scanning micromirror for LiDAR applications capable of 1.5 kHz and 4.4 kHz resonant frequencies with wide-angle optical deflection without external cooling. To overcome the limitations of conventional silicon-based MEMS scanners, a maskless 3D printing fabrication method using high-temperature polymer resin combined with Lorentz force electromagnetic actuation was adopted. This approach enables fast design iterations, simpler fabrication, lower costs, and application-specific customization, all while achieving commercial-grade LiDAR scanning mirror performance.
Two one-dimensional (1D) polymer MEMS scanners were designed and experimentally optimized for their actuation characteristics. The low-frequency scanner incorporates a 4.2 × 15 mm mirror with semicircular ends, achieving 34° peak-to-peak optical deflection at a 1.5 kHz resonant frequency. The high-frequency scanner employs a 4 mm circular mirror, producing over 6° optical deflection at 4.4 kHz resonance. Through orthogonal integration of these two scanners, a 2D scanning system was realized, providing a 34° horizontal field of view and approximately 8° vertical field of view, thereby satisfying the key requirements for robotics and short-range autonomous driving applications.
For optimized scanning device fabrication, methodologies were presented for deriving correlations between key parameters through analytical models and finite element analysis, as well as automated optimal geometry exploration through exhaustive search algorithm-based software. Through comprehensive analysis including printing materials, temperature-dependent material properties via tensile testing and dynamic mechanical analysis (DMA), process errors according to build orientation, anisotropy, and surface analysis, along with measured Young's modulus, the error between finite element analysis (FEA) and analytical models was reduced from approximately 30% to 8%, thereby establishing modeling reliability for polymer scanners.
In the 1.5 kHz scanning micromirror, thermal isolation anchors and thermal dissipation hole arrays were introduced to suppress resonance degradation due to Joule heating. Through these design modifications, when input current increased from 10 mArms to 170 mArms, the resonant frequency reduction was limited to less than 250 Hz while maintaining linear increase in optical scan angle. Particularly, to resolve the discrepancy between experimental and design values in the dynamic characteristics of the 4.4 kHz scanning micromirror, an anchor loss and stress mitigation model was developed through lateral clamping and perforated support structure design, minimizing the error between FEA simulation and experimental results to 0.34%.
An analytical method for Lissajous trajectory generation was investigated. Optimized scanner frequency pairs and phase difference combinations were pre-calculated and numerically verified, then compared against actual scanning patterns for validation. Additionally, a method to amplify the limited field of view of high-frequency scanners during system operation was proposed and demonstrated by aligning an additional scanner to the 2D module. The scanning micromirror developed in this research satisfies all key requirements for industrial robotics and short-range autonomous driving utilized in the Physical AI era and presents the practical commercialization potential of polymer-based LiDAR.

국문 초록 (Abstract)

본 연구는 외부 냉각 없이 1.5 kHz 및 4.4 kHz급 공진주파수와 광각 광학 편향이 가능한 3D 프린팅 기반 LiDAR용 스캐닝 마이크로미러를 개발하였다. 기존 실리콘 기반 MEMS 스캐너를 극복하기 위해, 로렌츠 힘 전자기 구동과 결합된 고온 폴리머 수지를 사용한 마스크리스 3D 프린팅 제조방식을 채택하였다. 이를 통해 빠른 설계, 제작 용이성, 경제성 및 응용별 맞춤 설정이 가능하며, 상용화 수준의 LiDAR용 스캐닝 미러 성능을 달성하였다. 두 개의 1D 폴리머 MEMS 스캐너를 설계하고 구동 특성을 실험적으로 최적화하였다. 저속 스캐너는 4.2 × 15 mm 반원형 단부 미러로 1.5 kHz 공진에서 34° 피크 투 피크 광학 편향을 달성하였고, 고속 스캐너는 4 mm 원형 미러를 통해 4.4 kHz에서 6° 이상의 광학 편향을 확보하였다. 두 폴리머 스캐너를 직교 배치하여 2D 스캐닝 시스템을 구현하였으며, 최대 34°의 수평 FOV와 최대 8°급 수직 FOV를 제공함으로써 로보틱스와 근거리 자율주행 시스템의 핵심 요구사항을 만족한다.
최적화된 스캐닝 소자 제작을 위해 해석적 모델과 유한요소해석을 통한 주요 파라미터 간 상관관계 도출 방법과 전수 검사 알고리즘 기반 소프트웨어를 통한 최적 형상 탐색 자동화 방안을 제시하였다. 프린팅 재료, 인장시험 및 DMA를 통한 온도 의존적 재료 특성, 출력 방향에 따른 공정 오차, 이방성, 표면 분석 등 포괄적인 분석과 측정한 영률(Young’s modulus)을 기반으로FEA와 해석 모델 간 오차를 약 30%에서 8%로 개선함으로써 폴리머 스캐너에 대한 모델링 신뢰성을 확보하였다. 저속 스캐너에서는 줄(Joule)열에 따른 공진 저하를 억제하기 위해 열 차단 앵커와 열 분산 홀 어레이를 도입하였다. 이러한 설계 변경을 통해 입력 전류가 10 mArms에서 170 mArms로 증가할 때 공진 주파수 감소폭을 250 Hz 미만으로 제한하면서도 광학 스캔 각도의 선형 증가를 유지하였다. 특히, 고주파 스캐너의 동적 특성 실험치가 설계치와 불일치하던 문제를 해결하기 위해, 측면 체결과 천공된 지지부 설계를 통한 앵커 손실 완화 모델을 개발하였고, FEA 시뮬레이션과 실험치 간 오차를 0.34 %로 최소화하였다. 리사쥬 궤적 생성의 해석적 방법을 조사하고, 최적화된 스캐너 주파수 쌍과 위상차 조합을 사전 계산하여 수치적으로 검증하였으며, 실제 패턴과의 비교 검증을 수행하였다. 또한, 2D 모듈에 추가 스캐너 하나를 정렬하는 방법을 통해 고주파 대역 스캐너의 제한적인 FOV를 시스템 구동 중 증폭할 수 있는 방안을 제시하고 시연하였다.
본 연구에서 개발한 스캐닝 마이크로미러는 Physical AI 시대에 활용되는 산업용 로보틱스 및 근거리 자율주행의 주요 요구사항을 모두 충족하며, 폴리머 기반 LiDAR의 실용적 상용화 가능성을 제시한다.

목차 (Table of Contents)

• Abstract １
• Chapter 1. Introduction １５
• 1.1 Overview of LiDAR and Applications in Robotics １５
• 1.1.1 Light Detection and Ranging System １５
• 1.1.2 LiDAR for Physical AI: Toward MEMS-Based Miniaturization １７
• 1.1.3 Application-Driven Specifications for Autonomous Robots ２１
• 1.2 Conventional LiDAR Systems ２８
• 1.2.1 Components of LiDAR system ２８
• 1.2.2 Laser Source ３１
• 1.2.3 Beam Steering ３３
• 1.3 MEMS Micromirrors ３８
• 1.3.1 Concept and Function ３８
• 1.3.2 Actuation Mechanisms ３９
• 1.3.3 Electromagnetic (EM) Scanners ４２
• 1.3.4 Application Driven Design Consideration ４９
• 1.4. 3D Printed Polymer Electromagnetic Scanning Micromirrors ５５
• 1.4.1 Limitations of Conventional Silicon MEMS Micromirrors ５５
• 1.4.2 Additive Manufacturing (AM) Approach ５８
• 1.4.3 Material feasibility ６０
• 1.4.5 Cost Efficiency and Total Cost of Ownership (TCO) ６３
• 1.4.6 Summary ６８
• Chapter 2. Research Objectives and Originality ６９
• 2.1. Research Objectives ６９
• 2.2. Originality ７１
• 2.3. Thesis Organization ７３
• Chapter 3. 1.5 kHz Scanning Micromirror with Structural Layer Material Characterization ７６
• 3.1. Introduction ７６
• 3.2. Design ７８
• 3.2.1 Architecture and operating principle ７８
• 3.2.2 Mirror plate geometry and optical layout ８０
• 3.2.4 Dynamics of torsional MEMS scanners ８１
• 3.2.5 Analytical model and parameter dependence ８６
• 3.2.5 Optimization software ９３
• 3.2.6 Finite element analysis １０５
• 3.2.7 Stress analysis １１１
• 3.2.8 Magnetic circuit design １１３
• 3.3. Fabrication １１６
• 3.3.1 Overview １１６
• 3.3.2 3D Printing Method and Material １１７
• 3.3.3 Structural Components １２２
• 3.3.4 Process Tolerances and Printing Orientation １２６
• 3.4. Device Characterization １３２
• 3.4.1 Driving Characteristics １３２
• 3.4.2 Structural-Layer Material Characterization １３８
• 3.4.3 Surface Topology １４５
• 3.4.4 Improvement of thermal properties １５０
• 3.4.5 Long-term Reliability and Package-level Drop Test １５７
• 3.5 Discussion １５９
• 3.6. Summary １６６
• Chapter 4. 4.4 kHz Scanning Micromirror Overcoming High-Frequency Challenges １６９
• 4.1. Introduction １６９
• 4.2. Design １７１
• 4.2.1 Design １７１
• 4.2.2 Analytical Modeling １７２
• 4.2.3 Finite Element Analysis (FEA) １７４
• 4.2.4 Anchor Loss Considerations １７７
• 4.3. Fabrication １８１
• 4.4. Device Characterization and High Frequency Effects １８５
• 4.4.1. Driving characteristics １８５
• 4.4.2 Thermal Effects and Polymer Modulus Temperature Dependence １８８
• 4.4.3 Anchor Loss Mitigation １９２
• 4.5. Summary １９３
• Chapter 5. 2D Lissajous Scanning System Using Orthogonal Polymer Scanners １９５
• 5.1. Introduction １９５
• 5.2. Fabrication of Orthogonal 2D Scanning System １９７
• 5.2.1 Architecture and Configuration １９７
• 5.2.2 Assembly and Experimental Setup １９９
• 5.3. Lissajous Scan Trajectory Fundamentals ２０３
• 5.3.1 Laser Scanning Patterns: Raster vs. Lissajous ２０３
• 5.3.2 Parametric Equations of Motion of Lissajous Scan ２０６
• 5.3.3 Frame Rate, Angular Resolution, Detection Range and Periodicity ２０７
• 5.3.3 Fill Factor and Coverage ２１６
• 5.3.4 Effects of Frequency Ratios and Phase ２１６
• 5.4. Device Characterization ２１８
• 5.4.1 Generated 2D Lissajous Scan Pattern ２１８
• 5.5. Lissajous Scan Optimization via Frequency-Phase LUT ２２５
• 5.5.1 Analytical studies on Lissajous scanning ２２５
• 5.5.2 Motivation for LUT-based approach ２２９
• 5.5.3 Performance Metrics of Lissajous Trajectory ２３１
• 5.5.3 Computational Search Framework and Implementation ２３３
• 5.5.4. Computational Analysis ２３８
• 5.5.5 Experimental Results ２４７
• 5.6. FOV Enhancement via Serial Scanner Integration ２５７
• 5.6.1 Concept and Motivation ２５７
• 5.6.2 Experimental Implementation and Results ２５９
• 5.7. Discussion and Summary ２６３
• Chapter 6. Summary and Conclusion ２６７
• 6.1 Summary ２６７
• 6.2 Conclusion and Future Work ２７２
• Bibliography ２７５
• Abstract in Korean ２８０
• Appendix ２８２
• A. Steel-Plate-Guided Polymer Scanner ２８２
• A.1 Introduction ２８２
• A.2 Design and Fabrication ２８３
• A.3 Discussion and Summary ２９０
• B. Fully 3D printed Scanner ２９２
• B.1 Introduction ２９２
• B.2 Design and Fabrication ２９２
• B.3 Discussion and Summary ２９８
• C. Exploiting 3D Printing for Advanced Spring Geometry and Dynamics ３０２