Baidu USA LLC

Microsoft Technology Licensing, LLC

Abbott Cardiovascular Systems, Inc.

Baidu.com Times Technology (beijing) Co. Ltd.

Dong Li

Yifei Jiang

Jiangtao Hu

Jiaming Tao

Liangliang Zhang

Guang Yu Yang

Jingao Wang

Fan Zhu

Qi Kong

Liyun Li

Qi Luo

Jinghao Miao

Yajia Zhang

Shuang Wu

Zhuang Li

Hongyi Sun

Kecheng Xu

Haoyang Fan

Li Zhuang

Youling Zou

Lei Ma

Xuedong D Huang

Jinyun Zhou

Shuangyu Chang

Li Jiang

Xiang Yu

Weiying Zhao

Xiangquan Xiao

Mudit Goel

Siyang Yu

Hemant Malhotra

Weizhen Wang

Yi Feng

Deping Xie

Ivo Jose Garcia Dos Santos

Jingtao Hu

Guang Yang

A sensor aggregation framework for autonomous driving vehicles is disclosed. In one embodiment, sensor data is collected from one or more sensors mounted on an autonomous driving vehicle (ADV) while the ADV is moving within a region of interest (ROI) that includes a number of obstacles. The sensor data includes obstacle information of the obstacles and vehicle data of the ADV. Each of the vehicle data is timestamped with a current time at which the vehicle data is captured to generate a number of timestamps that correspond to the vehicle data. The obstacle information, the vehicle data, and the corresponding timestamps are aggregated into training data. The training data is used to train a set of parameters that is subsequently utilized to predict at least in part future obstacle behaviors and vehicle movement of the ADV.

Sensor aggregation framework for autonomous driving vehicles

Embodiments are provided for securing data access to machine learning training data at a plurality of distributed computing devices. Electronic content including original data that corresponds to a preferred data security level is divided into a plurality of microsegments. The plurality of microsegments is restrictively distributed to a plurality of computing devices which apply transcription labels to the plurality of microsegments. The labeled microsegments are reconstructed into training data which is then used to train a machine learning model while facilitating an improvement in data security of the original data included with the training data from the reconstructed microsegments.

Microsegment secure speech transcription

According to one embodiment, a first phrase having one or more keywords is received to be added to a phrase blacklist, where the first phrase has been determined to be related to unwanted content. An analysis is performed on the keywords of the first phrase to identifying a set of one or more related phrases that are related to the first phrase. The first phrase and the set of related phrases are added in the phrase blacklist. The phrase blacklist contains a plurality of phrases that have been determined to be related to unwanted content. The phrase blacklist is utilized to prevent a content item from appearing in a search result, in response to a search query having a phrase matching at least one of the phrases in the phrase backlist.

Method and system for generating phrase blacklist to prevent certain content from appearing in a search result in response to search queries

According to various embodiments, systems and methods described in the disclosure combine mapped features with point cloud features to improve object detection precision of an autonomous driving vehicle (ADV). The map features and the point cloud features can be extracted from a perception area of the ADV within a particular angle view at each driving cycle based on a position of the ADV. The map features and the point cloud features can be concatenated and provided to a neutral network for object detections.

Comprehensive and efficient method to incorporate map features for object detection with LiDAR

An ADV may determine whether there is preexisting map data for an environment or geographical area/location where the ADV is located/travelling. If there is no preexisting data, the ADV may generate map data based on sensor data obtained from one or more sensors of the ADV. The ADV may determine a path for the ADV based on the generated map data. If there is preexisting map data, the ADV may determine a path for the ADV based on the preexisting map data.

Determining driving paths for autonomous driving vehicles based on map data

According to some embodiments, described herein is a system and method for handling sensor failures in autonomous driving vehicles (ADV) that is navigating in a world coordination as an absolute coordination system. When the ADV encounters a sensor failure, but still has at least one camera working properly, the sensor failure handling system can switch the ADV from navigating in the world coordination to a local coordination, in which the ADV relies camera-based obstacle detection and lane mark detection to drive safely until human dis-engagement or until the ADV is parked along a road side.

Methods for handling sensor failures in autonomous driving vehicles

A perception module is configured to perceive a driving environment surrounding an autonomous driving vehicle (ADV) based on sensor data, and to generate perception information using various perception models or methods. The perception information describes the perceived driving environment. Based on the perception information, a planning module is configured to plan a trajectory representing a route or a path for a current planning cycle. The ADV is then controlled and driven based on the trajectory. In addition, the planning module determines a critical region (also referred to as a critical area) surrounding the ADV based on the trajectory in view of a current location or position of the ADV. The metadata describing the critical region is transmitted to the perception module via an application programming interface (API) to allow the perception module to generate perception information for a next planning cycle in view of the critical region.

Perception and planning collaboration framework for autonomous driving

Via a first processing thread, an ADV is controlled according to a first trajectory that was generated based on a first reference line starting at a first location. Concurrently via a second processing thread, a second reference line is generated based on a second location of the first trajectory that the ADV will likely reach within a predetermined period of time in future. The predetermined period of time is greater than or equals to an amount of time to generate a reference line for the ADV. The second reference line is generated while the ADV is moving according to the first trajectory and before reaching the second location. Subsequently, in response to determining that the ADV is within a predetermined proximity of the second location, a second trajectory is generated based on the second reference line without having to calculate the second reference line at the second location.

Method and system for generating reference lines for autonomous driving vehicles using multiple threads

A server may determine whether traffic control devices in an environment have changed, based on one or more report messages and sensor data. If the traffic control devices in an environment have changed, the server may generate updated map data and transmit the updated map data to an autonomous driving vehicle (ADV).

Updating map data for autonomous driving vehicles based on sensor data

System and method for training a machine learning model are disclosed. In one embodiment, for each of the driving scenarios, responsive to sensor data from one or more sensors of a vehicle and the driving scenario, driving statistics and environment data of the vehicle are collected while the vehicle is driven by a human driver in accordance with the driving scenario. Upon completion of the driving scenario, the driver is requested to select a label for the completed driving scenario and the selected label is stored responsive to the driver selection. Features are extracted from the driving statistics and the environment data based on predetermined criteria. The extracted features include some of the driving statistics and some of the environment data collected at the different points in time during the driving scenario.

System and method for training a machine learning model deployed on a simulation platform

In response to a search query received from a client, a content item is identified from a content database based on one or more keywords associated with the search query. In addition, a search is performed in an image store to identify a list of image candidates based on one or more keywords associated with the search query. An image is selected from the list of image candidates using an image selection algorithm. Metadata associated with the content provider is inscribed onto the selected image to generate a customized image. The customized image is integrated with the content item to generate a composite content item such as a poster. As a result, the content represented by the composite content item is closely tied to the content provider. The composite content item is then transmitted to the client.

Method and system for dynamically overlay content provider information on images matched with content items in response to search queries

In one embodiment, a system monitors states of an autonomous driving vehicle (ADV) using a number of sensors mounted on the ADV. The system perceives a driving environment surrounding the ADV using at least a portion of the sensors. The system analyzes the states in view of the driving environment to determine a real-time traffic condition at a point in time. The system determines whether the real-time traffic condition of the driving environment matches at least a predetermined traffic condition. The system transmits data concerning the real-time traffic condition to a remote server over a network to allow the remote server to generate an updated map having real-time traffic information, in response to determining the real-time traffic condition is unknown. In response to receiving the updated map, the system plans and controls the ADV based on the updated map.

System for building a vehicle-to-cloud real-time traffic map for autonomous driving vehicles (ADVS)

A social driving style learning framework or system for autonomous vehicles is utilized, which can dynamically learn the social driving styles from surrounding vehicles and adopt the driving style as needed. Each of the autonomous vehicles within a particular driving area is equipped with the driving style learning system to perceive the driving behaviors of the surrounding vehicles to derive a set of driving style elements. Each autonomous vehicle transmits the driving style elements to a centralized remote server. The server aggregates the driving style elements collected from the autonomous vehicles to determine a driving style corresponding to that particular driving area. The server transmits the driving style back to each of the autonomous vehicles. The autonomous vehicles can then decide whether to adopt the driving style, for example, to follow the traffic flow with the rest of the vehicles nearby.

Group driving style learning framework for autonomous vehicles

An autonomous driving vehicle (ADV) may determine a predicted path for a moving obstacle and speeds for different portions of the path. The ADV use multiple threads in parallel to determine the path and speeds for the different portions of the path.

Determining driving paths for autonomous driving using parallel processing

In some implementations, a method is provided. The method includes determining a path for an autonomous driving vehicle. The path is located within a first lane of an environment in which the autonomous driving vehicle is currently located. The method also includes obtaining sensor data. The sensor data indicates a set of speeds for a set of moving obstacles located in a second lane of the environment and wherein the second lane is adjacent to the first lane. The method further includes determining whether the set of speeds is lower than a threshold speed. The method further includes determining a new speed for the autonomous driving vehicle in response to determining that the set of speeds is lower than the threshold speed. The method further includes controlling the autonomous driving vehicle based on the path and the new speed.

Growing community of inventors

Sunnyvale, CA, United States of America

Dong Li