CSCE Capstone
Student Site for Individual and Collaborative Activites
Team 8 – NASA Robotic Mining Competition
Team Members:
Project Summary:
The goal of this project is to design, implement, and evaluate the computer systems of a robotic lunar mining rover developed by the Razorbotz team at the University of Arkansas. In this project, the old robot code will be refactored and documented with good programming practices and a newer platform, while expanded with new autonomy features. Over several months, the team will first refactor the old code, document the updated code, and then expand upon it. The goal is to win the NASA Robotics Mining Competition (RMC) in May 2021 with the updated robot and control system. In the future, lunar mining robots can be used to extract minerals on the moon, reducing costs of such materials on Earth while eliminating the environmental impacts from their extraction.
Task List
Updated 4/29/2021
Legend:
- ✅ : Completed
- 🕒 : Future Work
- Refactor and Upgrade Old Code
- ✅ Update ROS1 code to ROS2 (William)
- ✅ Trim Current Modules (William)
- Study current code to understand functionality
- Remove code that is no longer applicable to the current project
- Communicate these specifics with design teams
- ✅ Create/Update Documentation (Jett)
- Long-lasting and extremely detailed discussions of each ROS2 node
- ✅ Create Dockerfile (Gunner)
- Allows for easy development off the robot computer
- ✅ Update ROS1 code to ROS2 (William)
- Improve Manual Control and UI
- ✅ Update UI to be compatible with current robot
- 🕒Create a user-friendly user interface (William)
- Implement Full Autonomy
- Excavation Autonomy
- 🕒 Design, implement, and train the model
- 🕒 Simulate the model on the rover
- 🕒 If computation power permits, run the trained model directly on the rover
- Dump Autonomy
- ✅ Use object recognition to recognize the mining base
- 🕒 Create analysis algorithm that determines the correct path to the dump site
- 🕒 Create functions to control motors to dump rocks
- Travel Autonomy
- ✅ Implement YOLO object recognition neural network (Carson)
- ✅ Build dataset of rocks and hazards (Gunner)
- ✅ Integrate ZED into the ROS2 system (Calvin)
- 🕒 Train YOLO to recognize rocks and hazards using the dataset
- 🕒 Create analysis algorithm that parses recognized objects and hazards to calculate the optimal path to the next rock
- Failure Management
- 🕒 Detect any component failures, such as camera or accelerometer loss
- 🕒 Implement backup features for detectable losses
- Excavation Autonomy
- Win the NASA RMC Competition
- ✅ Complete design document (All members)
- 🕒 Compete in 2022
Task Schedule
Updated 4/28/21
|
Tasks |
Target Dates |
|
GUI/CLI Design Started |
9/1/2020 |
|
GUI Design Review |
9/26/2020 |
|
GUI Programming/Code Rewrite Started |
10/1/2020 |
|
Start Manual Controls |
10/1/2020 |
|
Preliminary Design Review |
10/10/2020 |
|
Finish GUI/CLI Design Overhaul |
12/18/2020 |
|
Revisit Scope of Project |
1/16/2020 |
|
Start Excavation Macro |
1/23/2021 |
|
Start Camera Implementation |
1/23/2021 |
|
Start Sensor Implementation |
1/23/2021 |
| Start Training AI and Dataset Creation |
2/6/2021 |
|
Complete Target Localization with ZED Camera |
2/27/2021 |
| Finish Manual Controls Overhaul | 3/20/2021 |
| Begin Testing Camera and Sensors While Driving | 4/30/2021 |
| Begin Full Scale Mining Testing | 4/30/2021 |
| Graduate | 5/7/2021 |
|
Compete |
|
<iframe width=”560″ height=”315″ src=”https://www.youtube.com/embed/vs-5AqC1drE” title=”YouTube video player” frameborder=”0″ allow=”accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture” allowfullscreen></iframe>
Contact:
Uche Wejinya
Links
Proposal Documents:
Final Documents:
Source Code:
