Feasibility to convert ROS based LiDAR object detection code to Driveworks format using Point Cloud Processing API

Autonomous Vehicles DRIVE AGX Xavier DRIVE AGX Xavier General
1

Please provide the following info (check/uncheck the boxes after creating this topic):
Software Version
DRIVE OS Linux 5.2.6
DRIVE OS Linux 5.2.6 and DriveWorks 4.0
DRIVE OS Linux 5.2.0
[*] DRIVE OS Linux 5.2.0 and DriveWorks 3.5
NVIDIA DRIVE™ Software 10.0 (Linux)
NVIDIA DRIVE™ Software 9.0 (Linux)
other DRIVE OS version
other

Target Operating System
[*] Linux
QNX
other

Hardware Platform
NVIDIA DRIVE™ AGX Xavier DevKit (E3550)
[*] NVIDIA DRIVE™ AGX Pegasus DevKit (E3550)
other

SDK Manager Version
[*] 1.7.1.8928
other

Host Machine Version
[*] native Ubuntu 18.04
other

Is it feasible to convert a ROS based LiDAR object detection code with publishers and subscribers sending actuation signals based on live LiDAR point cloud data into Driveworks format using Point Cloud Processing API?

3

Dear @raji,
Does that mean, you are looking for integration of DW + ROS code? Or you have lidar recorded data using ROS and want to use in DW code?

4

Dear @SivaRamaKrishnaNV ,

Does that mean, you are looking for integration of DW + ROS code?

We don’t want to reinvent the wheel and with minimal changes to ROS based env code to work with Driveworks API.

Or you have lidar recorded data using ROS and want to use in DW code?

We want our algorithm to process live lidar data and send actuation signals based on the algorithm’s detection.

Please let me know if I’m unable to understand your question correctly.

Our algorithm is working for live LiDAR data and also sending actuation signals to DBW vehicle. We want to replicate the same thing using Driveworks.

So how should we proceed in general?

  1. if we want to convert the ROS environment based code to Driveworks environment then do we need to change the code completely?
  2. Can Point cloud processing API process the live lidar data then perform object detection/segmentation and send actuation with the help VehicleIO module?

Please confirm if my understanding is correct.

5

Dear @raji,

  1. if we want to convert the ROS environment based code to Driveworks environment then do we need to change the code completely?

There is no publishers/ subscribers mechanism in DW. You need to use SAL APIs to read sensor data. You may check Lidar samples to know about API usage.

  1. Can Point cloud processing API process the live lidar data then perform object detection/segmentation and send actuation with the help VehicleIO module?

Point cloud processing API can perform operations given at DriveWorks SDK Reference: Point Cloud Processing on lidar data. We don’t have object detection algorithm APIs based on lidar data. You may have to implement object detction DNN model based on lidar data and use DW DNN module to integrate into DW.