velodyne reader

This repo provides python scripts to read data stream directly from velodyne lidars.

Prerequisite

conda create -n lidar --file requirements.txt -c conda-forge
conda activate lidar
pip install 'velodyne_decoder>=3.0.0'

Or, if you prefer pip

pip install -r requirements.txt
pip install 'velodyne_decoder>=3.0.0'

For velodyne_decoder, please make sure you install >=3.0.0. If velodyne_decoder>=3.0.0 is not available via pip, install from source at develop branch.

pip install git+https://github.com/valgur/velodyne_decoder.git@develop

Applicability

Generally, the supported model is decided by valgur/velodyne_decoder. You can get the supported device list by

$ python -c "import velodyne_decoder as vd;print(list(vd.Model.__entries))"
['HDL64E_S1', 'HDL64E_S2', 'HDL64E_S3', 'HDL32E', 'VLP32A', 'VLP32B', 'VLP32C', 'VLP16', 'PuckLite', 'PuckHiRes', 'VLS128', 'AlphaPrime']

Usage

Independently

Take VLP-16 for example, the following command runs a 60-second demo, and write the point cloud to out/<date>/<model>_<rpm><returnMode>_<UTCtimestamp>.pcap, together with cooresponding .log file. The .pcap file can be visualized directly by VeloView.

python vldreader.py --model VLP-16 --ip-lidar <ip_addr_lidar> --dataport <data_port_lidar> --rpm <rpm>

If you did not tune the lidar's ip address and its data port, simply run

python vldreader.py --model VLP-16 --rpm 1200

As a component

Of course, you can also use this script as a module in your project.

from /path/to/vldreader.py import ld

myld=ld(model='VLP-16', lidarip='192.168.1.201', dataPort=2368, rpm=600, returnMode='dual')

# launch the device
myld.launch()

# read stream data
for onepkt in myld.read_live_data():
    print(onepkt)

# stop the device
myld.stop()