Data collection

1 Preparation

Note

Exp1 and Exp2 are mandatory, Exp3 and Exp4 are optional since they can be done offline as well.

Time:
- Dec. 13
- 9:00 AM - 9:00 PM
TODO:
- For each laptop, we need to record their own rosbag.
- Run roscore on laptop1 and laptop2 separately and redo the first TODO.
- Each lidar needs to connect with GPS and synchronized.
Goal:
- To mimic that we are running data on each laptop separately.
- Find out how big is the data in rosbag, according to previous, the total rosbag is about 13.3GB/ 63s
- Test the limitation of our power source
Scenarios:
- Morning / Afternoon / Night
- Each rosbag is about 1 min
- Test rosbag is about 20s, conduct at least 3 test rosbag to verify everything works.
- For each scenarios, collect at least 3 set of data, 3 mins in between.
Actions:
- Record rosbag on each laptop at the same time, requires two people
- Record the start time of power source (connected to both Lidar and GPS)

Time:
- Same as exp 1
TODO:
- Collect some test data to adjust the tf parameters.
- Record the mapping overall rosbag on laptop1
- Also record unmapping rosbag on laptop2 as comparison data
- As a comparison, record the mapping overall rosbag on laptop2 and unmapping rosbag on laptop 1
- Record mapping on both laptop1 and laptop2
Goal:
- To have a view of combined early fusion map.
- Verify the difference between different data through put
Actions:
- Run ucla_mapping to fix the tf between two lidars.

Time:
- Dec. 14 and Dec. 15
- 9 AM - 9 PM
TODO:
- Setup the detection algorithm on at least one laptop
- Use online detection algorithm to detect the rosbag on both laptop1 and laptop2 separately.
Goal:
- Verify the online detection algorithm
- Verify the difference between different data throughput
Actions:
- Run online 3D detection algorithm on laptop1