ME336

This is the projects folder for course ME336, which is based on DeepClaw. The projects consist of experimental learning contents from week06 to week16. We will release new contents in this folder before the class, so please visit the repository to get updated. The task to be completed in each week is listed below. Please refer to the subfolders for more detailed instructions. Supplementary data and pretrained model can be downloaded from Baidu disk:

链接: https://pan.baidu.com/s/1YlTTVqXheK-27PvgyhG9ZA 提取码: pzdr

DeepClaw (fork from WangHK backbone)

DesignAIR_simulation (GeS)

The subfolder DesignAIR_simulation[./DesignAIR_simulation] contains all the source codes for robot simulations.

Week 01~05 The Missing Semester by MIT (By LiuXB)

Week 06 Introduction to Python (By LiuXB)

The programming language you are going to use for ME336 is python. This week you are required to preparing the environment on your computer by installing all the required softwares. The guidelines are included in the subfolder week06.

The learning goals are listed below:

• Install Linux
• Practice Python programming skills

Week 07 Introduction to TensorFlow (By LiuXB)

The deep learning framework adopted by ME336 is tensorflow from Google. This week you are required to install tensorflow on your own computer and learn the basic usages. The installation guidelines and learning resources are included in the subfolder week07.

The learning goals are listed below:

• Install tensorflow 2.x
• Finish the basic usage tutorial on image learning.

Week 08 Introduction to PyRep (GeS)

Before making hands-on pratices with the real robots, the students will learn how to build a robot and environment and complete certain tasks in simulation. This is very useful and safe when you first learn how to program a robot when you don't have access to the real robot or you are new to the robot hardware. ME336 adopts Pyrep, which is the python programming interface build on top of CoppeliaSim (previously called V-REP). The installation guidelines and learning resources are included in the subfolder week08.

In this week, the students are required to run a simple robot picking task using PyRep.

The learning goals are listed below:

• Install PyRep
• Follow the tutorial to complete the kinematic Picking project.

Week 09 Simulated Robot Player (GeS)

• Program the TicTacToe game in PyRep simulation.
• Provides complete video, codes and a 5-page report for the project.

Week 10 Simple Waste Classifier (WanF)

• Train a simple waste classifier based on AlexNet.

Week 11 Introduction to DeepClaw and Claw Machine (GeS+LiuXB)

• Read the instructions on how to use DesignAir
• Complete the Arcade Claw machine tutorial

Week 12 Waste Sorting Optimization (GeS+LiuXB)

• Download and get familiar with the recycling trash dataset
• Finetune a detection model in tensorflow to classify and localize the trash.

Week 13 Waste Sorting Robot (GeS+LiuXB)

• Test the detection model on hardware with Franka and realsense

Week 14: Interactive DesignAIR (GeS)

Week 15: Final Presentation (LiuXB)