Minghao Ning

CoInfra cooperative infrastructure perception with globally fused road-user tracks at a real traffic site.

Minghao Ning

Ph.D. Candidate, University of Waterloo

Autonomous driving, robotics, perception, sensor fusion, motion planning, and cooperative infrastructure autonomy.

I am a Ph.D. candidate in Mechanical and Mechatronics Engineering at the University of Waterloo, advised by Prof. Amir Khajepour and co-supervised by Prof. Ehsan Hashemi. My research focuses on building deployable perception and planning systems for autonomous vehicles and mobile robots, especially under adverse weather, occlusion, network latency, and real-world deployment constraints.

Research Focus

Perception under difficult sensing conditions LiDAR, camera, and HD map fusion for drivable-space estimation, calibration, tracking, SLAM, and uncertainty-aware scene understanding.

Infrastructure-assisted autonomy Roadside sensor networks, V2X sensing, multi-node synchronization, delay-aware fusion, and globally consistent traffic perception.

Planning and control on real platforms Point-cloud potential fields, MPC trajectory generation, collision avoidance, vehicle/robot control, and closed-loop validation.

Edge-cloud connected mobility 5G/Wi-Fi communication, edge perception, cloud-level fusion, safety warnings, connected vehicles, and indoor mobile robots.

Selected Systems and Project Highlights

Fielded autonomous vehicle

WATonoBus autonomous shuttle

Led perception development for a full-scale autonomous shuttle, including LiDAR-camera-HD map drivable-space estimation and field-test validation in complex road scenes and Canadian winter conditions. Also led the transition from a custom in-house autonomy stack toward an Autoware-based next-generation platform.

Read paper View platform

Supporting links: drivable-space code, UWaterloo News, MathWorks, ESV.

CoInfra outdoor cooperative infrastructure perception with globally consistent 3D tracking.

Roadside infrastructure network

CoInfra cooperative infrastructure perception

Designed and deployed a 14-node roadside testbed with cameras, LiDAR, edge computers, and 5G/Wi-Fi for adverse-weather traffic perception. The system supports calibration, synchronization, cloud-side fusion, globally consistent 3D tracking, and public data/code release.

Read paper Code/dataset

Supporting link: arXiv preprint.

Edge-cloud cooperative mobility system integrating infrastructure sensing and connected autonomy.

Edge-cloud autonomy framework

Cloud-based Autonomous Mobility (CAM)

Developed an edge-cloud mobility framework that connects infrastructure sensing to autonomy services for shared scene understanding, safety warnings, connected autonomous driving, and robot mobility across outdoor and indoor environments.

Read paper Related ITSC paper

Supporting link: arXiv preprint.

Indoor cooperative perception supporting closed-loop MPC on a medical-bed robot platform.

Indoor robot mobility

Indoor cooperative perception and robot mobility

Built ceiling-mounted cooperative sensing for occlusion-heavy indoor spaces, including uncertainty-aware 3D human pose estimation and delay-aware tracking. The perception output feeds closed-loop MPC on a real medical-bed robot platform.

SAP-CoPE paper Indoor paper

Supporting links: SAP-CoPE code, indoor code, UWaterloo News.

Selected Publications

CoInfra: A Large-Scale Cooperative Infrastructure Perception System and Dataset in Adverse Weather. Under review at Transportation Research Part C, 2026. [project] [arXiv] [code]
Real-World Deployment of Cloud Autonomous Mobility System Using 5G Networks for Outdoor and Indoor Environments. Under review at IEEE Intelligent Transportation Systems Magazine, 2026. [project] [arXiv]
SAP-CoPE: Social-Aware Planning using Cooperative Pose Estimation with Infrastructure Sensor Nodes. Revised manuscript under review at IEEE Transactions on Automation Science and Engineering, 2026. [project] [arXiv] [code]
A Novel Motion Planning for Autonomous Vehicles Using Point Cloud Based Potential Field. IEEE Transactions on Vehicular Technology, 2024. [project] [DOI]
An Efficient Approach to Generate Safe Drivable Space by LiDAR-Camera-HD Map Fusion. IEEE International Conference on Intelligent Transportation Systems, 2024. [project] [DOI] [code]
Enhancing Indoor Mobility with Connected Sensor Nodes: A Real-Time, Delay-Aware Cooperative Perception Approach. IEEE International Conference on Intelligent Transportation Systems, 2024. [project] [DOI] [code]

See the full Publications page and my Google Scholar profile for the complete and most up-to-date publication list.