Course Center/"Learning by Doing" AI Practice – Grades 7–9 (Hardware)
"Learning by Doing" AI Practice – Grades 7–9 (Hardware)
Middle School24 lessons
Course Overview
This course follows the main line of “Build — Program — Create with Intelligence,” guiding students to understand intelligent system design methods through mechanical structure building and interactive programming practice, and to develop engineering practice ability and computational thinking through hands-on creation. In Grade 7, students learn mechanical structure building and interactive programming, build models such as a catapult, a gearbox, and a rocket car, understand gear transmission and linkage principles, and write interactive game programs. In Grade 8, they learn combined applications of mechanical transmission and program control, build models such as a night light, a bug robot, and a spinning-top launcher, and complete interactive game design using conditionals, logical operations, and variable management. In Grade 9, they learn AI robot dog building and programming, master servo control and motion programming, build agent workflows on the Coze platform for voice interaction, and use AI tools to assist in generating robot dog programs.
Included Courses
Grade 7
8 class periods
Grade 8
8 class periods
Grade 9
8 class periods
Competency Goals
Self-Directed Learning
Actively explore how mechanical structures and AI robots work, discover and solve problems independently in building and programming practice; use AI tools to look up materials for learning, and adopt AI-generated results with care.
Independent Decision-Making
When choosing mechanical solutions, designing gear ratios, and designing program logic, collect information for comparison and analysis, and make reasonable choices based on independent judgment.
Self-Management
Independently plan progress for model building, program debugging, and project optimization; allocate class time reasonably; record each test result and adjust the plan accordingly.
Engineering Thinking
Understand the engineering flow of intelligent systems—“structure building → mechanical transmission → program control → test and debug”—and evaluate and optimize solutions under constraints such as part quantity and hardware performance.
Computational Thinking
Decompose interactive programming tasks into modules such as character control, logic judgment, and data management; use variables, conditions, loops, and other programming structures for abstract modeling; analyze and fix program errors.
Design Thinking
Design interactive games and smart robot motions from user experience, iteratively improve function and experience based on test feedback, and create works that are both fun and practical.
Cultural Understanding
In projects such as the catapult and biomimetic flight, understand the fusion of technology with history and nature, recognize how technological development affects social life, and think about ways technology can serve society.
Communication & Collaboration
Express ideas effectively and collaborate with division of labor in group building and testing; clearly describe program design ideas, listen to others, and complete works together.
Critical Thinking & Innovation
Identify anomalies in program running and limitations of mechanical design, independently evaluate their reasonableness; propose personally creative improvements based on existing solutions.
Course Advantages
Real Training, Real Experience
Students personally build physical models such as a catapult, a gearbox, and an AI robot dog, write programs to control mechanical motion, and understand how intelligent systems work through the full real process of “structure building → mechanical transmission → program control → test and debug.”
Stepwise Progression
From mechanical structure building and interactive programming, to combining mechanical transmission with program control, then to AI robot dogs and agent applications—three stages rise in a spiral so beginners can keep up step by step.
Hardware–Software Collaboration
On top of hardware building, students write programs to control motors, sensors, and actuators, integrating physical structure, mechanical transmission, and program logic to create interactive works that can actually run.
AI Robot Dog in Practice
In Grade 9, students use a real AI robot dog, build agent workflows on the Coze platform to control hardware with voice commands, experience the core agent chain of “sense → decide → act,” and encounter frontier AI applications early.
