ROS Resources: Documentation | Support | Discussion Forum | Index | Service Status | ros @ Robotics Stack Exchange
Ask Your Question
1

CUDA dependency in autoware.auto

asked 2020-07-26 21:36:06 -0500

Sharoon gravatar image

Hi,

I have a question with respect to the CUDA dependency in autoware.auto. Does Autoware.auto do not have any CUDA dependencies ?

I do not see any CUDA files (.cu) in the autoware.auto version of the software

Could you please let me know that, whether CUDA dependency is completed removed from all modules of autoware.auto software ?

Is autoware.auto does not use CUDA programming model for parallel processing ?

edit retag flag offensive close merge delete

Comments

Hi @Sharoon, I come up with the same question, did you find the answer?

runtao gravatar image runtao  ( 2021-01-10 16:18:02 -0500 )edit

2 Answers

Sort by ยป oldest newest most voted
0

answered 2021-05-31 05:45:47 -0500

Josh Whitley gravatar image

Autoware.Auto does not depend on CUDA. We will be adding machine-learning components but the back-ends which will be supported first are Vulkan and OpenCL due to them being more open and hardware-agnostic than CUDA. Please see https://gitlab.com/autowarefoundation... for tracking progress.

edit flag offensive delete link more
0

answered 2021-05-31 16:13:26 -0500

liyou zhou gravatar image

For parallel processing:

there is only currently a limited amount of code in Autoware.auto that requires parallel acceleration. Previously OpenMP has been used https://gitlab.com/autowarefoundation.... There has not been a need for massive parallel acceleration in Autoware just yet. In the future when a need arises, we will choose an appropriate API to use.

Avoiding CUDA is deliberate because it is a proprietary API with vendor hardware lock-in. This is highly undesirable in an open-source project such as Autoware.

For Machine Learning:

We are currently working on adding machine learning facilities in Autoware via TVM https://gitlab.com/autowarefoundation.... This allows the abstraction of backends. We have chosen libvulkan for our reference implementation. This means there is no vendor lockin for hardware acceleration.

edit flag offensive delete link more

Question Tools

2 followers

Stats

Asked: 2020-07-26 21:36:06 -0500

Seen: 452 times

Last updated: May 31 '21