Mingming Zhang 张明明 (hitcm)-Alibaba AI Labs

I am currently leading the robotics team as staff engineer and tech manager in Alibaba, in charging of Alibaba's exploration in Embodied AI, covering the entire chain of Data-Model-Hardware-Applications, and integrating the capabilities of Qwen LLM with robots.

Prior to this, I completed a six-year advanced program at Harbin Institute of Technology, culminating in a Master of Robotics Engineering awarded by State Key Laboratory of Robotics and System in 2016.

During my college life, I successfully attained the following achievements:

1. Placed 2nd among over 100 teams in a national robotics competition named Robocon
2. Pioneered dynamic mobility control & real-time navigation stack for quadruped robots, inspired by Boston Dynamics' BigDog legacy.

During my Alibaba tenure, I successfully delivered these core projects:

1. Mobile manipulator with Franka(we named it as "MOFA")
   1. Leveraging the Qwen multimodal large language model, we perform domain-specific fine-tuning with robotics-oriented datasets to enhance spatial perception capabilities. The optimized model achieves edge deployment on Jetson mobile platforms, serving as an intelligent agent that performs high-level cognitive functions including environmental perception, task planning, and human-robot interaction.
   2. VLA (Vision-Language-Action) model generates end-effector poses for subsequent low-level control execution. This enables diverse specialized tasks in indoor environments including door handle manipulation, elevator operation, switch actuation, and valve control.
   3. Implementing reinforcement learning for whole-body control optimization, leveraging base motion to accelerate arm manipulation.
2. XiaoManLv or “Little Donkey”
   1. An L4 autonomous delivery vehicle, engineered to resolve final three-kilometer delivery challenges for express parcels, food deliveries, and fresh groceries. Its core mission is enabling efficient, low-cost unmanned logistics through self-driving technology.
   2. XiaoManLv surpasses 1,000 autonomous delivery units in global deployment milestone.
3. Last-mile Delivery Robot named “Antman”
   1. The robot retrieves parcels at residential community entrances, navigates internal pathways to apartment buildings, autonomously operates elevators, and delivers packages directly to customer doorsteps.
   2. Near-hundred-unit robotic fleet deployment achieved.

I have also published 10+ high-quality papers in top robotics conferences, ICRA, IROS and TRO,etc

My research is focusing on robotics, specifically, perception, planning and manipulation etc.

Feel free to contact me if you are interested in amazing robotics positions in Alibaba.

Github: https://github.com/hitcm

E-mail: robotsming@gmail.com

Google Scholar: G+

Projects and Publications

Closing the Loop: Real-Time Visual-Language-Action Control for Mobile Dual-Arm Systems Mingming Zhang, etc 2025

Mobile manipulation Couples Base and Arm Motion, Boosting Productivity Through Unified Motion Control. Mingming Zhang, etc 2024

Franka Omni-Manipulation System: Autonomous Goods-Packing, Door-Opening, Elevator-Riding and Knob-Actuation. Mingming Zhang, etc 2024

3D Terrain-Adaptive Locomotion and Navigation for Quadruped Robots. Mingming Zhang, etc 2023

XiaoManLv Surpasses 1,000 Autonomous Delivery Units in Global Deployment Milestone Mingming Zhang, etc 2023 [ news, news2]

3D Reconstruction for Static Scene and Dynamic Objects With Online Images. Mingming Zhang, etc 2022 [ video]

All-weather and All-terrain Robotic Applications. Mingming Zhang 2022 [ video]

High Precision Repeatability for Robotic Localization and Planning. Mingming Zhang 2021 [ video]

LIDAR Based Multi Object Tracking. Mingming Zhang 2021 [ video]

LIDAR Odometry, Mapping and Relocalization in Pre-built Map Mingming Zhang 2020 [ video]

Multi Agent Visual Inertial Mapping Mingming Zhang 2020 [ video]

Kinematics Based Visual Localization for Skid-Steering Robots: Algorithm and Theory Xingxing Zuo, Mingming Zhang, Yiming Chen, Guoquan Huang, Yong Liu, Mingyang Li Preprint in arXiv [ pdf ]

Localization for Ground Robots: On Manifold Representation, Integration, Re-Parameterization, and Optimization Mingming Zhang, Xingxing Zuo, Yiming Chen and Mingyang Li Preprint in arXiv [ pdf, video]

Monocular Visual-inertial Odometry for Different Mobile Platforms. Mingming Zhang 2018 [ video]

Real-time Pose and Depth Estimation on Mobile Devices Mingming Zhang 2020 [ video]

Real-time Pose and Depth Estimation on Mobile Devices Mingming Zhang 2020 [ video]

Visual-Inertial Localization for Skid-Steering Robots with Kinematic Constraints Xingxing Zuo, Mingming Zhang, Yiming Chen, Yong liu, Guoquan Huang, and Mingyang Li ISRR 2019 [ pdf ]

Perception System Design for Low-Cost Commercial Ground Robots: Sensor Configurations, Calibration, Localization and Mapping Yiming Chen, Mingming Zhang, Dongsheng Hong, Chengcheng Deng, and Mingyang Li IROS 2019 [ pdf]

Vision-Aided Localization For Ground Robots Mingming Zhang, Yiming Chen and Mingyang Li IROS 2019 [ pdf, video]

SDF-Loc: Signed Distance Field based 2D Relocalization and Map Update in Dynamic Environments Mingming Zhang, Yiming Chen and Mingyang Li ACC 2019 [ pdf, video]

Visual Inertial Map Alignment and Bundle Adjustment Mingming Zhang 2019 [ video]

Real-time Stereo Direct Sparse Odometry Without Loop closure Mingming Zhang 2017 [ video]

Mobile Phone AR Recognition and Tracking Mingming Zhang 2017 [ video]