公开(公告)号：US09417069B2

公开(公告)日：2016-08-16

申请号：US13951015

申请日：2013-07-25

Applicant: HONDA MOTOR CO., LTD.

Inventor： Rakesh Gupta ,  Igor V. Karpov ,  Antoine Raux ,  Deepak Ramachandran

IPC: G06F7/60 ,  G06F17/10 ,  G01C21/00 ,  G01C21/34 ,  G01C21/36 ,  G06N99/00

CPC classification number: G01C21/00 ,  G01C21/3484 ,  G01C21/3641 ,  G06N99/005

Abstract: One or more embodiments of techniques or systems for modeling familiarity for a traveler are provided herein. Familiarity evidence can be received, indicative of how familiar a traveler is with an area or road segment, and based on a number of visits the traveler has made to that area. The familiarity evidence can be used to generate one or more familiarity models indicative of a predicted familiarity of locations around the area. Familiarity models can be based on kernels, graph distances, Markov random fields (MRFs), etc. When route directions are generated from an origin location to a destination location, one or more of the directions can be provided based on one or more of the familiarity models. For example, if a familiarity model indicates that a traveler is familiar with a route, driving directions of the route can be adapted to be more succinct.

Abstract translation: 本文提供了用于建模旅客的熟悉度的技术或系统的一个或多个实施例。 可以接收到熟悉的证据，指示旅行者在区域或路段中的熟悉程度，以及旅客对该地区的访问次数。 熟悉证据可用于产生一个或多个熟悉模型，表明该地区周边地区的熟悉程度。 熟悉度模型可以基于内核，图形距离，马尔可夫随机场（MRFs）等。当从原点位置到目标位置生成路线方向时，可以基于一个或多个方向来提供一个或多个方向 熟悉模式。 例如，如果熟悉模式表明旅行者熟悉路线，路线的驾驶方向可以更加简洁。

公开(公告)号：US20140361973A1

公开(公告)日：2014-12-11

申请号：US13934396

申请日：2013-07-03

Applicant: Ian Lane ,  Honda Motor Co., Ltd.

Inventor： Antoine Raux ,  Ian Lane ,  Rakesh Gupta

IPC: G06F3/16 ,  G06F3/01

CPC classification number: G06F3/167 ,  G01C21/3629 ,  G01C21/3664 ,  G06F1/1694 ,  G06F3/01 ,  G06F3/013 ,  G06F3/017 ,  G06F2203/0381

Abstract: A method and system for multimodal human-vehicle interaction including receiving input from an occupant in a vehicle via more than one mode and performing multimodal recognition of the input. The method also includes augmenting at least one recognition hypothesis based on at least one visual point of interest and determining a belief state of the occupant's intent based on the recognition hypothesis. The method further includes selecting an action to take based on the determined belief state.

Abstract translation: 一种用于多模式人车交互的方法和系统，包括经由多于一种模式从车辆中的乘员接收输入并执行输入的多模式识别。 该方法还包括基于至少一个视觉兴趣点增加至少一个识别假设，并且基于识别假设确定乘客意图的置信状态。 该方法还包括基于所确定的信念状态来选择要采取的动作。

公开(公告)号：US09857795B2

公开(公告)日：2018-01-02

申请号：US15079917

申请日：2016-03-24

Applicant: HONDA MOTOR CO., LTD.

Inventor： Rakesh Gupta ,  Robert Cofield

IPC: G05D1/02 ,  G05D1/00

CPC classification number: B60W30/20 ,  B60W2520/10 ,  B60W2550/10 ,  B60W2550/402

Abstract: A trajectory planning system for an autonomous vehicle fits a jerk profile including a plurality of phases within a set of acceptable parameters, a jerk value being constant within each phase of the jerk profile. The system parameterizes the jerk profile based on an initial velocity, an initial acceleration, a final velocity, and a final acceleration for the first segment. The system then integrates the jerk profile to determine a first trajectory function, the first trajectory function including a speed for the vehicle at a given time. The system guides the vehicle along the first segment of the path according to the first trajectory function. The system detects an unplanned obstacle along the first segment of the path. The system plans a second trajectory function for a second segment of the path between a current location of the vehicle and a location on the path before the unplanned obstacle.

公开(公告)号：US09784592B2

公开(公告)日：2017-10-10

申请号：US14801904

申请日：2015-07-17

Applicant: Honda Motor Co., Ltd.

Inventor： Rakesh Gupta ,  Bo Tang ,  Salman Anwar Khokhar

IPC: G01C21/00 ,  G01C21/36 ,  B60W30/00

CPC classification number: G01C21/3697 ,  B60W30/00

Abstract: One or more embodiments of techniques or systems for generating turn predictions or predictions are provided herein. Environment layout information of an operating environment through which a first vehicle is travelling may be received. A current location of the first vehicle may be received. One or more other vehicles may be detected. Additional environment layout information from other vehicles may be received. A model including the operating environment, the first vehicle, and one or more of the other vehicles may be built. The model may be based on the environment layout information and the additional environment layout information and indicative of an intent of a driver of one of the other vehicles. Further, predictions may be generated based on the model, which may be based on a Hidden Markov Model (HMM), a Support Vector Machine (SVM), a Dynamic Bayesian Network (DBN), or a combination thereof.

公开(公告)号：US20170016734A1

公开(公告)日：2017-01-19

申请号：US14801904

申请日：2015-07-17

Applicant: Honda Motor Co., Ltd.

Inventor： Rakesh Gupta ,  Bo Tang ,  Salman Anwar Khokhar

IPC: G01C21/34

CPC classification number: G01C21/3697 ,  B60W30/00

Abstract: One or more embodiments of techniques or systems for generating turn predictions or predictions are provided herein. Environment layout information of an operating environment through which a first vehicle is travelling may be received. A current location of the first vehicle may be received. One or more other vehicles may be detected. Additional environment layout information from other vehicles may be received. A model including the operating environment, the first vehicle, and one or more of the other vehicles may be built. The model may be based on the environment layout information and the additional environment layout information and indicative of an intent of a driver of one of the other vehicles. Further, predictions may be generated based on the model, which may be based on a Hidden Markov Model (HMM), a Support Vector Machine (SVM), a Dynamic Bayesian Network (DBN), or a combination thereof.

Abstract translation: 本文提供了用于产生回合预测或预测的技术或系统的一个或多个实施例。 可以接收第一车辆行驶的操作环境的环境布局信息。 可以接收第一车辆的当前位置。 可以检测到一个或多个其他车辆。 可以接收来自其他车辆的附加环境布局信息。 可以构建包括操作环境，第一车辆和一个或多个其他车辆的模型。 该模型可以基于环境布局信息和附加环境布局信息并且指示其他车辆之一的驾驶员的意图。 此外，可以基于可以基于隐马尔可夫模型（HMM），支持向量机（SVM），动态贝叶斯网络（DBN）或其组合的模型来生成预测。

公开(公告)号：US09653071B2

公开(公告)日：2017-05-16

申请号：US14465890

申请日：2014-08-22

Applicant: Honda Motor Co., Ltd.

Inventor： Rakesh Gupta ,  Teruhisa Misu ,  Aasish Pappu

IPC: G10L15/22 ,  G10L15/08 ,  G10L15/14 ,  G10L15/32

CPC classification number: G10L15/14 ,  G10L15/32

Abstract: A method and system are disclosed for recognizing speech errors, such as in a spoken short messages, using an audio input device to receive an utterance of a short message, using an automated speech recognition module to generate a text sentence corresponding to the utterance, generating an N-best list of predicted error sequences for the text sentence using a linear-chain conditional random field (CRF) module, where each word of the text sentence is assigned a label in each of the predicted error sequences, and each label is assigned a probability score. The predicted error sequence labels are rescored using a metacost matrix module, the best rescored error sequence from the N-best list of predicted error sequences is selected using a Recognition Output Voting Error Reduction (ROVER) module, and a dialog action is executed by a dialog action module based on the best rescored error sequence and the dialog action policy.

公开(公告)号：US09286029B2

公开(公告)日：2016-03-15

申请号：US13934396

申请日：2013-07-03

Applicant: Honda Motor Co., Ltd. ,  Ian Lane

Inventor： Antoine Raux ,  Ian Lane ,  Rakesh Gupta

IPC: G09G5/00 ,  G06F3/16 ,  G06F3/01

CPC classification number: G06F3/167 ,  G01C21/3629 ,  G01C21/3664 ,  G06F1/1694 ,  G06F3/01 ,  G06F3/013 ,  G06F3/017 ,  G06F2203/0381

Abstract: A method and system for multimodal human-vehicle interaction including receiving input from an occupant in a vehicle via more than one mode and performing multimodal recognition of the input. The method also includes augmenting at least one recognition hypothesis based on at least one visual point of interest and determining a belief state of the occupant's intent based on the recognition hypothesis. The method further includes selecting an action to take based on the determined belief state.

Abstract translation: 一种用于多模式人车交互的方法和系统，包括经由多于一种模式从车辆中的乘员接收输入并执行输入的多模式识别。 该方法还包括基于至少一个视觉兴趣点增加至少一个识别假设，并且基于识别假设确定乘客意图的置信状态。 该方法还包括基于所确定的信念状态来选择要采取的动作。

公开(公告)号：US20150228272A1

公开(公告)日：2015-08-13

申请号：US14465890

申请日：2014-08-22

Applicant: Honda Motor Co., Ltd.

Inventor： Rakesh Gupta ,  Teruhisa Misu ,  Aasish Pappu

IPC: G10L15/01 ,  G10L15/26

CPC classification number: G10L15/14 ,  G10L15/32

Abstract: A method and system are disclosed for recognizing speech errors, such as in a spoken short messages, using an audio input device to receive an utterance of a short message, using an automated speech recognition module to generate a text sentence corresponding to the utterance, generating an N-best list of predicted error sequences for the text sentence using a linear-chain conditional random field (CRF) module, where each word of the text sentence is assigned a label in each of the predicted error sequences, and each label is assigned a probability score. The predicted error sequence labels are rescored using a metacost matrix module, the best rescored error sequence from the N-best list of predicted error sequences is selected using a Recognition Output Voting Error Reduction (ROVER) module, and a dialog action is executed by a dialog action module based on the best rescored error sequence and the dialog action policy.

Abstract translation: 公开了一种用于识别语音错误的方法和系统，例如在口语短消息中，使用音频输入设备接收短消息的话语，使用自动语音识别模块来生成对应于话语的文本语句，生成 使用线性链条件随机场（CRF）模块的文本句子的预测误差序列的N最佳列表，其中在每个预测误差序列中分配文本句子的每个单词的标签，并且分配每个标签 概率得分。 使用Metacost矩阵模块对预测的错误序列标签进行重新索引，使用识别输出投票错误减少（ROVER）模块选择来自预测错误序列的N最佳列表的最佳重排错误序列，并且由 对话框动作模块，基于最佳重击错误顺序和对话框动作策略。