I am a PhD student in the Robotics Institute at the University of Michigan, USA. My advisor is Maani Ghaffari and I'm in Computational Autonomy and Robotics Laboratory. I received my Master's degree from UM Robotics. Prior to that, I was a Research Assitant at National Taiwan University where I earned my Mechanical Engineering Bachelor degree. My research interest is perception in robotics. My current research topic is visual odometry.

email: chienerh at umich dot edu

[CV][Google Scholar]


A New Framework for Registration of Semantic Point Clouds from Stereo and RGB-D Cameras

A novel nonparametric rigid point cloud registration framework that jointly integrates geometric and semantic measurements such as color or semantic labels into the alignment process and does not require explicit data association.

Mini Cheetah Perception

Using inEKF for state estimation and adding camera and lidar to perform semantic mapping on mini cheetah.


Used SIFT-flow as adiitional labels to improve Contineous Visual Odometry (CVO).

Computer Vision-Based Navigation System for Visually Impaired Person

Developed a computer vision-based system that can navigate someone visually impaired to the targeted door with object detection, object tracking, and visual SLAM.

Structured Light 3D Reconstruction

Reconstructed an underwater scene in 3D, using structured light systems that project light patterns on the objects and enable the reconstruction of its structure, despite the low texture and low contrast of underwater images.

Pest Surveillance System

Constructed an IoT-based wireless imaging and sensor node system for remote greenhouse pest monitoring, with 90% accuracy of pest classification using deep neural network, enabling farmers easily notice amount change of pest in real-time.

Plant Growth Monitoring System

Built a Plant Growth Monitoring System to monitor the growth of Cordyceps militaris using Qt, OpenCV and Raspberry Pi.


"IE2-8: A Real-time Multi-class Insect Pest Identification Method using Cascaded Convolutional Neural Networks." (Dan Jeric Arcega Rustia, Chien Erh Lin, and Jui-yung Chung), In 9th International Symposium on Machinery and Mechatronics for Agriculture and Biosystems Engineering (ISMAB), 23, no. 1 (2018): 67-67. [Publisher link]

"Pest surveillance system." (Dan Jeric Arcega Rustia, Chien Erh Lin, and Ta-Te Lin.), U.S. Patent Application 15/990,791, filed July 25, 2019. [Patent Link]

"Innovation of New Occlusion Devices for Cancers." (Hao-Ming Hsiao, Tzu-Yuan Lin, Chien-Erh Lin, Han-Yu Lee, and Yi-Ping Wang), Applied Sciences 7, no. 5 (2017): 530. [Publisher link]

"A Novel Spherical Stent Concept for Intracranial Aneurysm." (Hao-Ming Hsiao, Yi-Ping Wang, Yu-Han Cheng, Tzu-Yuan Lin, and Chien-Erh Lin), Sensors and Materials 28, no. 9 (2016): 947-955. [Publisher link]

"A Novel Spherical Stent for Occlusion of Cancer and Aneurysm." (Hao-Ming Hsiao, Wen-Hsin Yang, Tzu-Yuan Lin, Chien-Erh Lin, and Jiong-Hong Chen), In the Biomedical Engineering Society (BMES) Annual Meeting, Phoenix, USA, October 11-14, 2017.

"Drug-eluting stent with rhombic-shape reservoirs for drug delivery." (Yen-Ting Wang, Yi-Ping Wang, Tzu-Yuan Lin, Chien-Erh Lin, and Hao-Ming Hsiao), In In 2016 International Conference on Applied System Innovation (ICASI), pp. 1-4. IEEE, 2016. [BibTeX][Publisher link]


Image Caption Generator with Simple Semantic Segmentation

Utilized a pre-trained ImageNet as the encoder, and a Long-Short Term Memory (LSTM) net with attention module as the decoder in PyTorch that can automatically generate properly formed English sentences of the inputted images and achieved BLEU-4 score 0.2320 (out of 1) with beam search size 5 in evaluation. Implemented a simple semantic segmentation algorithm using the sentence generated along with it’s attention layer.

Evaluation of LeGO-LOAM

Evaluated LeGO-LOAM (Lightweight andGround-Optimized Lidar Odometry and Mapping) which reduced computational expense while keeping similar accuracy compared to LOAM method using KITTI Odometry Benchmark Dataset, UTBM Dataset and KAIST Urban Dataset. Compared the mapping and odometry results of these data, and analysis the relative motion and mapping error in this report.

Vehicle Classification and Localization

Utilizing OpenCV for image preprocessing and TensorFlow for DenseNet deep learning to classify 22 different types of vehicle and localize them by point cloud data.

Mobile Robot with Particle Filter and Path Planning

Implementing a waypoint navigation system, a simultaneous localization and mapping (SLAM) system using 2D LiDAR, an A* path planning algorithm, and a visualization of the environment for a mobile robot.


NA/EECS 568, ROB 530: Mobile Robotics: Methods and Algorithms -- Winter 2021, GSI

This course is a graduate-level topic on the theory and application of probabilistic techniques for autonomous mobile robotics.

EECS 565: Linear Feedback Control Systems -- Winter 2019, Grader

This courses builds on undergraduate frequency domain methods and graduate-level state-variable methods in order to develop feedback design concepts for linear multivariable systems.


UM Discover Engineering

Helped design and organize a robotics coding activity for a two day camp for local high school students. Guided students in the workshop through the robotics line following and grasping task. The camp focused on increasing their interest in STEM, as well as over-viewing its academic and career paths.

Ann Arbor Summer Festival KidZone

Explained the theory and the usage of Lidar to local families.