| | 89 | |
| | 90 | == Week 9 Activities == |
| | 91 | |
| | 92 | * Switched to ROSARIA, now using ROS nodes to connect to and wirelessly drive the robot |
| | 93 | * Configured ROS nodes such that training data consisting of both imaging and robot commands can |
| | 94 | be recorded simultaneously |
| | 95 | * Began collecting data through the rosbag node |
| | 96 | * Properly explored the gazebo simulation, expanding on models and the world |
| | 97 | * [https://docs.google.com/presentation/d/1KjHGXeC8n4A_bMgDGMsrH517wKPCVwpcG06br4tpleE/edit?usp=sharing Week 9 Presentation] |
| | 98 | |
| | 99 | == Week 10 Activities == |
| | 100 | |
| | 101 | * Debugged Python files to reflect Vector3 messages that we used to represent steering data |
| | 102 | * Trained robot on circular track to learn how to properly steer to avoid going off-road |
| | 103 | * Subscribed to control and image topics and recorded training using rosbag |
| | 104 | * Converted bag files to image (.ndz) files that were fed into neural network with 4 convolutional |
| | 105 | layers |
| | 106 | * [https://docs.google.com/presentation/d/1iZCfFvj8ZyDHKDPy_X8Xqb8ji-_jtblDsoBADHxNKZ0/edit?usp=sharing Final Presentation] |
| | 107 | |
