Reconstructing, Representing, and Understanding Human Mobility.
What can we recover, represent, and reliably infer from incomplete mobility data?
I develop computational methods that bridge noisy observations, interpretable mobility representations, and transportation planning, advancing our understanding of human movement and the limits of mobility data.
How much can we trust what we observe?
- Sparse & noisy GPS / LBS traces
- Sampling gaps across modes
- Measurement uncertainty
What can incomplete mobility observations reliably reveal about human movement?
- Mobility recoverability
- Trip & activity inference
- Planning & transit accessibility
About Me
Hi! I'm Ruohan! My research focuses on human mobility modeling, mobility data analytics, and computational methods. I enjoy developing algorithms such as Sup-HMM, PM-Tree, and HSMM-T, and building NovaMatch, an integrated framework for large-scale mobility data processing, map matching, network representation, and analytics.
I aim to bridge methodological research with real-world transportation challenges. Through projects supported by FHWA/USDOT, VDOT, DVRPC, and PennDOT, I have contributed to applied mobility research and transportation planning. My experience also includes real-world airport traffic simulation and analysis using VISSIM.
My work includes two first-author publications in IEEE Transactions on Intelligent Transportation Systems, with five additional first-author manuscripts under review. My doctoral research, From Observations to Mobility Representations, explores overlooked challenges and limitations in mobility data, with the goal of developing robust, interpretable, and trustworthy methods for transportation and urban analysis.
Selected Research
Representative work on mobility reconstruction, representation, and real-world transportation applications.
Sup-HMM
Bayesian-optimized map matching for sparse and noisy LBS trajectories.
IEEE Transactions on Intelligent Transportation Systems · 2026
PM-Tree
Priority-guided hierarchical representation of sparse trajectories.
IEEE T-ITS · Under Review
Complete Trip
A linked multimodal human mobility dataset reconstructed from passive LBS observations.
TRB 2026 / AGU25 · 2026
Transit Mirage
Rethinking the interpretation of transit demand-supply imbalance.
Transportation Research Part A · Under Revision
Latest Updates
- Oct 2026 Sup-HMM published in IEEE T-ITS (Early Access)!
- Sep 2026Three papers accepted for presentation at TRB 2027 (106th Annual Meeting).
- Jul 2026Complete Trip dataset released as a preprint.
- Oct 2025Presenting Transit Deserts work at TRB 2026, Lectern Session 4018.
- 2026: 🎉 My first paper from my Ph.D. journey, Taming Volatile Trajectories: Sup-HMM, a Bayesian-Optimized Hidden Markov Model for Map Matching of Location-Based Service Data, has been accepted as a regular paper by IEEE Transactions on Intelligent Transportation Systems (IEEE T-ITS) on 27-Sep-2026!
- 2027: Three of my papers have been accepted for presentation at the TRB 106th Annual Meeting! 🎉
- 07/2026: Our lab's latest dataset paper, Complete Trip: A Linked Multimodal Human Mobility Dataset, is now publicly available on arXiv. arXiv.
- 10/2025: I will present my paper Rethinking Transit Deserts: Identifying Pseudo and Real Gaps through Trip-Level Mobility Data at the TRB 105th Annual Meeting, as the first presenter in Lectern Session 4018: Evaluation of Non-Conventional Transit Impacts (Wednesday, Jan 14, 8:00-9:45 AM, Room 150B, Convention Center). Welcome to attend and join the discussion!
- 09/2025: My paper Rethinking Transit Deserts: Identifying Pseudo and Real Gaps through Trip-Level Mobility Data has been accepted by TRB. Another paper, Complete Trip: An Open and Privacy-Safe Dataset of Multimodal Travel Sequences for Urban Transportation Systems Analysis, has also been accepted.
- 08/2025: After the TRB deadline, I have been busy cleaning up my LBS map-matching code. Formatting code so that other humans can understand it feels like the hardest optimization problem ever.
- 09/2024: The Computer Software Copyright Registration in China: Vision-Based Lane Detection System With Self-supervised Pre-training Through Masked Sequential Auto-encoders (VLD_SP-MSAE), granted on Sep. 11, 2024, Patent number 2024R11L1180902. We also published the open-source lane detection code on GitHub.
- 09/2024: I presented my recent work, Sup-HMM Map-Matching of Location Data Trajectories: A Heterogeneous and Bayesian-Optimized Hidden Markov Approach, at the TRC-30 conference in Crete.
- 07/2024: My paper have been submitted and is currently under review by The Transportation Research Board (TRB) 104nd Annual Meeting: Sup-HMM Map-Matching of Location Data Trajectories: A Heterogeneous and Bayesian-Optimized Hidden Markov Approach. Working on the map-matching continues. This work is part of our team's Complete Trips Project.
- 06/2024: My paper has been accepted by Conference in Emerging Technologies in Transportation Systems (TRC-30): Map Matching of Location Data Trajectories: A Heterogeneous and Bayesian-Optimized Hidden Markov Approach. Very excited for the presentation in Crete, Greece, in September!
- 06/2024: The computer software has been submitted and is currently under review by Computer Software Copyright Registration in China: Lane Line Detection Software Based on Image Sequence Mask Pre-Training.
- 04/2024: My paper has been submitted and is currently under review by Conference in Emerging Technologies in Transportation Systems (TRC-30): Map Matching of Location Data Trajectories: A Heterogeneous and Bayesian-Optimized Hidden Markov Approach. I'm currently working on map matching for human mobility data.
- 10/2023: Two of my TRB papers were accepted: A Novel Highway Traffic Capacity Analyzing Method under Road Region Atmospheric Environment Constraints Based on Computational Fluid Dynamics Model (Rank 1st) and Intelligent Anomaly Detection for Lane Rendering Using Transformer with Self-Supervised Pre-Training and Customized Fine-Tuning (Rank 3rd), TRB 103rd Annual Meeting.
- 09/2023: The article, "The Highway Region Boundary Based on Multi-Environmental Factors," can now be accessed. View the article here.
- 09/2023: Starting my Ph.D. studies at Villanova University!
- 08/2023: The article, "Robust Lane Detection Through Self Pre-Training With Masked Sequential Autoencoders and Fine-Tuning With Customized PolyLoss," has been finalized and posted in the "Early Access" area on IEEE Xplore. View the article here.
- 07/2023: I received my Ph.D. offer from Villanova University! Looking forward to new research and life at Villanova!
- 07/2023: My paper Robust Lane Detection through Self Pre-training with Masked Sequential Autoencoders and Fine-tuning with Customized PolyLoss is accepted by the journal of IEEE Transactions on Intelligent Transportation Systems.
- 06/2023: I received my M.Eng. degree in traffic and transportation from Lanzhou Jiaotong University!