|
Work item:
|
H.VM-GAVCC
|
|
Subject/title:
|
Requirements and framework of generative AI enabled vehicle-cloud collaboration system for vehicular multimedia
|
|
Status:
|
Under study [Issued from previous study period]
|
|
Approval process:
|
AAP
|
|
Type of work item:
|
Recommendation
|
|
Version:
|
New
|
|
Equivalent number:
|
-
|
|
Timing:
|
2026-Q3 (Medium priority)
|
|
Liaison:
|
SG13, CITS
|
|
Supporting members:
|
Chongqing Changan Automobile
|
|
Summary:
|
Traditional vehicular AI interaction system of vehicular multimedia system is widely used for simple scenarios such as open the window, find the nearest gas station and so on. Currently, generative AI has reached a level of maturity suitable for complex vehicle scenarios. Due to limited computing resources on the vehicle side, it is necessary to implement cooperative processing between vehicle and cloud (vehicle-cloud collaboration (VCC)). This recommendation specifies generative AI based vehicle-cloud collaboration (GAVCC) system framework, use cases, vehicle-cloud API requirements and functional requirements, including vehicle-cloud intent recognition and response, multi-intention planning, personalized driver assistance, multimodal interaction information generation in vehicle, full-scenario agent about driving. GAVCC system based on this Recommendation improves user experience of driver and passengers.
|
|
Comment:
|
-
|
|
Reference(s):
|
|
|
Historic references:
|
|
Contact(s):
|
|
| ITU-T A.5 justification(s): |
|
|
|
|
First registration in the WP:
2024-09-18 16:20:40
|
|
Last update:
2024-09-18 17:22:42
|
