白丝美女被狂躁免费视频网站,500av导航大全精品,yw.193.cnc爆乳尤物未满,97se亚洲综合色区,аⅴ天堂中文在线网官网

System for in-vehicle-infotainment based on dual asynchronous displays

專利號(hào)
US11175876B1
公開(kāi)日期
2021-11-16
申請(qǐng)人
Ford Global Technologies, LLC(US MI Deerborn)
發(fā)明人
Fling Tseng; Aed M. Dudar; Jóhannes Geir Kristinsson
IPC分類
G06F3/14; B60K35/00; G06N3/04; G06N3/08
技術(shù)領(lǐng)域
adaptive,driver,may,system,rule,vehicle,or,display,computing,learning
地域: MI MI Dearborn

摘要

Multiple display infotainment systems in vehicles provide various options for drivers and passengers to interact with elements on the multiple displays. Techniques include accessing historical driver information that includes previous or typical driver preferences and actions while operating the vehicle. Driving context information and driver state information are received describing the conditions in and around the vehicle. When a signal indicates that an interactive element should be presented to the driver or a passenger, an adaptive rule-based system uses the contextual information and previous history to determine a score for each interaction element. The score and they type of interaction required by the element are then used to determine a location on the multiple displays for presenting the interactive element.

說(shuō)明書(shū)

In the depicted adaptive learning system 400, a forecasting model may be generated such that the hidden layer(s) 406 retains information (e.g., specific variable values and/or transformative functions) for a set of input values and output values used to train the adaptive learning system 400. This retained information may be applied to a new contextual data, such as the contextual data 302, in order to identify a likelihood of a particular outcome, the particular outcome representing the likelihood of driver interaction with an interactive element. In some embodiments, the adaptive learning system 400 may be trained on samples of historical driver data having known outcomes.

By way of illustration, an adaptive learning system as depicted in FIG. 4 may be trained using both known historical driver outcome data and context data as inputs. Each of the output nodes in this example may represent results positioned within a hyperspace. When a new contextual dataset is presented as input to the trained adaptive learning system, the adaptive learning system will output a result which can be assessed based on its position within the hyperspace.

權(quán)利要求

1
微信群二維碼
意見(jiàn)反饋