Versions Compared

Key

  • This line was added.
  • This line was removed.
  • Formatting was changed.

PTL : Liya Yu (13 April 2021)


Blueprint species:

 

Use Case Attributes

Description

Informational

Type

New

 

Blueprint Family

Integrated Edge Cloud (IEC)The AI Edge

 

Use CaseSafety and security, driving assistant

Autonomousdrivingtaxi

 

Blueprint proposed Name

IEC Type 3: Autonomous vehicles as the edgeRoboTaxi

 

Initial POD Cost (capex)

N
/A

Leverage Unicycle POD - less than $150k

 

Scale & Type

One per vehicle

Up to 4 servers, x86/Arm server or deep edge class

With nVIDIA Tesla P4/T4 GPUs

 

Applications

Video Processing for AutoDriving (especially training deep learning models)

  • lane guidance
  • safety and security applications
  • remote control

Autonomousdrivingtaxi

 

Power and memory restrictionsAutonoumous vehicles are power and memory constrained.

Less than 10Kw

 

Infrastructure orchestrationNeed Geolocation correlation with deep learning model (the trained ones in binary format)

Docker 1.13.1 or above

K8s 1.12.5 or above- Container Orchestration

OS – CentOS 7.0 or above

 

SDN

Calico

and K8s, and Containers serving as client/server (a database to collect raw videos/video streams) and server/client (a database for trained models in binary format)

container networking,or OVS-DPDK

 

Workload Type

Containers

(Tensoflow, Keras containers)VMs

 

Dataplane VPP w/ Direct GPU (e.g. GPUnet)

Additional Details

Different bus connection: Ethernet, CANbus, etc.

There are mainly 4 pieces:

  1. a database to collect raw videos/video streams
  2. tons of containers with various models on training
  3. a database for trained models in binary format
  4. a database for queries records and transactions of database mentioned in 1 and 3.
PaaSOTE Stack + Baetyl + OpenNESS

Additional Details

N/A

 



View file
nameRoboTaxi_V1.3.pdf
height250

As per the Akraino Community process and directed by TSC, any person wishing to be PTL shall self nominate themselves by putting a Y in the column.   A blueprint which has only one nominee for Project Technical Lead (PTL) will be the elected lead.   If there are two or more, an election will take place.

Self Nominations begin on 25 March 2021 and will conclude on 1 April 2021


Committer

Committer

Company

 Committer Contact Info

Committer Bio

Committer Picture

Self Nominate for PTL (Y/N)

Hechun Zhang

Baidu

zhanghechun@baidu.com

zhanghechun@gmail.com

Seasoned MEC and 5G Solution Architect of Baidu, with impressive 9 years  experience in the ICT industry.

Project Manager of ODCC Edge Computing Group.

张贺纯.jpegImage Added



Zhenhua XuBaiduxuzhenhua02@baidu.com


Liya YuBaiduyuliya@baidu.com

Y
Zhuming Zhang(Simmy Zhang)Yishi17555827@qq.com


Zhongwang Hao(haozw)

Didi

Arm
Yishizw.hao@msn.cn


Yipan DengIntelyipan.deng@intel.com


Haitao WangIntelhai.tao.wang@intel.com


Xiaopeng TongIntelxiaopeng.tong@intel.com


Tina TsouArmtina.tsou@arm.com


Dechao KongBaidukongdechao@baidu.com


Wenhui ZhangPSU




N



Proposed PTL: 

Ken YiHechun Zhang

Contributors: 

@soshun@soshun.arai@arm.com

Tina Tsou (Deactivated) 

Wenhui Zhang

Committers: 

Ken Yi 

Tina Tsou (Deactivated)