Chengdong ZhuRESEARCH & ENGINEERING

ROBOTICS · VISION · SYSTEMS

Chengdong Zhu

Robot integration. Perception, control & AI.

I am a PhD researcher in Electronic Engineering at the University of York. My areas of focus are robot system integration, robot perception, robot control and AI development.

University of York · York, United Kingdom

Front view of the V1 underwater research robot and its stereo cameras
ROV V1Stereo vision research platform
Areas of focusRobot system integrationRobot perceptionRobot controlAI development
01

Projects across four areas

What I built, how it works and what has been verified.

Two views of the assembled V1 underwater research robot

01 / Robot system integration

ROV V1

Integrating a robot platform for underwater vision experiments.

My work

  • Designed the mechanical structure and sensor mounts around a stereo camera payload.
  • Integrated electronics, wiring and software control interfaces into the research platform.
  • Organized supervised shallow-water tests and engineering documentation.

Built research platform · supervised shallow-water testing

View integration work
York-Tank experiment with an overhead camera and AprilTag reference array

02 / Robot perception

York-Tank

Connecting stereo data collection, visual degradation and SLAM evaluation.

My work

  • Integrated the tank experiment platform and carried out stereo camera calibration.
  • Collected and organized stereo sequences by degradation type and difficulty.
  • Carried out visual SLAM evaluation and prepared the dataset and paper.

Published dataset and paper · 16 sequences · 18,948 stereo pairs

View perception work
Existing flexible-arm platform photographed during collaborative vineyard field work
Existing project platform · vineyard field photograph
Real photograph of the flexible-arm drive area, motors and wiring
Drive-area photograph · actuator hardware and wiring

03 / Robot control

ATArm · 3D position control

Extending an existing flexible-arm controller with external visual position feedback.

My work

  • Developed local numerical-Jacobian control and alternating/coordinated extension–bending strategies.
  • Connected ROS pose feedback, coordinate transforms and measurement freshness checks.
  • Added actuator allocation, bounded updates and independent final-position checks; tested with simulation and fake serial interfaces.

Software prototype · offline validation complete · physical commissioning pending

View control methods and validation
Native SolidWorks preview of the generated bracket, shaft and plate assembly
Generated assembly · native SolidWorks isometric preview
SolidWorks AI rc13 workbench showing a completed preview review task
Actual workbench · requirement, plan and review status

04 / AI development

SolidWorks AI

Turning model planning and tool calls into an inspectable CAD workflow.

My work

  • Built model-provider adapters and a workbench for structured part and assembly planning.
  • Connected MCP tools to serial native SolidWorks API execution, with staged saving and revision tracking.
  • Integrated geometry checks and preview review; packaged native CAD, STEP, BOM and a Windows application.

Windows prerelease available · native CAD validated on SolidWorks 2024

View the software and download
02

Publication & open resources

03

Research background

Engineering across vision and robotics.

My background combines mechanical engineering with doctoral research in Electronic Engineering. I am interested in how visual perception methods behave when they meet the practical constraints of robotic experiments.

Download public CV
  1. University of York

    PhD researcher · Electronic Engineering

  2. China University of Mining and Technology

    MEng · Mechanical Engineering

  3. China University of Mining and Technology

    BEng · Mechanical Engineering

GET IN TOUCH

Research, engineering and collaboration.

For research collaboration and robotics R&D enquiries, please contact me by email.