Beijing Baidu Netcom Science and Technology Co., Ltd.

Apollo Intelligent Driving Technology (beijing) Co., Ltd.

Baidu Online Network Technology (beijing) Co., Ltd.

Chengliang Deng

Zhuo Chen

Kaiwen Feng

Bocong Liu

Wei He

Yibing Liang

Yu Ma

Liming Xia

Haifeng Wang

Jiankang Xin

Weifeng Yao

Chao Zhang

Quan Wang

Ning Qu

Jingchao Feng

Zhou Chen

Weide Zhang

Zhongjun Ni

The embodiments of the present disclosure provide a method, an apparatus, a device and a medium for processing topological relation of tasks. The method includes: extracting at least one execution element from each of processing tasks based on a topological relation recognition rule; determining a dependency relation among the processing tasks according to content of the execution element of each processing task; and determining a topological relation among the processing tasks according to the dependency relation among the processing tasks.

Method, apparatus, device and medium for processing topological relation of tasks

The present disclosure provides a method and an apparatus for scheduling a task, a device and a medium. The method includes: obtaining a processing task to be executed from a task queue via a main thread bound to a processor, in which the processor is configured to execute the main thread, to execute the processing task; obtaining a newly triggered processing task; in response to determining that a priority of the newly triggered processing task is greater than or equal to a priority of the processing task executed on the main thread, assigning the newly triggered processing task to a standby thread; and dispatching and assigning a processor from at least one processor through the system kernel to execute the standby thread.

Method and apparatus for scheduling task, device and medium

Embodiments of the present disclosure provide a method, a device and a vehicle for message deduplication. The method includes: determining a node relationship with each publishing node by a subscribing node based on property information of the publishing node under the same topic as the subscribing node; determining a communication mode with each publishing node according to the node relationship; registering management modules corresponding to different communication modes according to the communication modes, so that after receiving a service message from the publishing node, a management module determines which subscribing nodes the service message is transmitted to based on a registration content and a unique identification of the publishing node carried in the service message. The service message may be transmitted to the corresponding subscribing node based on the registration content specifying the publishing node from which the service message is received under a certain communication mode.

Method, device and vehicle for message deduplication

The present application discloses a method and an apparatus for transmitting a message. The method of an embodiment comprises: dividing a shared memory area allocated by an operating system into a plurality of memory blocks for storing messages having different priorities; dividing each memory block in the plurality of memory blocks into at least one memory unit having a successive serial number; determining, from the plurality of memory blocks according to a priority of a to-be-transmitted message, a target memory block for storing the to-be-transmitted message; and writing the to-be-transmitted message into a memory unit of the target memory block. This embodiment implements message transmission based on priorities of the messages, and improves the control security of an intelligent control system.

Method and apparatus for transmitting a message

The present application discloses a data processing method and apparatus. A specific implementation of the method includes: simultaneously acquiring, by a data processing party, data to be used by a plurality of data receiving parties, and processing the to-be-used data according to processing mode information that is received in advance from each data receiving party and that indicates a processing mode for the to-be-used data, to obtain processed to-be-used data corresponding to the data receiving party; and storing the processed to-be-used data into a storage area corresponding to the data receiving party. An independent data processing party for simultaneously acquiring data that needs to be processed by each data receiving party is provided, so that the data that needs to be processed by each data receiving party is acquired at a time.

Data processing method and apparatus

The present application discloses a data acquisition method and apparatus for a driverless vehicle. The driverless vehicle is provided with a LIDAR sensor and a camera sensor, and the method of an embodiment comprises: acquiring a collection period of the LIDAR sensor as a first period; acquiring a start time of a current time window; executing following data processing steps: executing real-time acquiring and storing operations on data packets collected by the LIDAR sensor and the camera sensor after the start time of the current time window; and determining whether a following condition is met: the first period has elapsed from the start time of the current time window to a current time; setting the start time of the current time window to the current time in response to determining that the condition is met, and continuing executing the data processing steps. This embodiment realizes synchronized storage of data packets collected by the LIDAR sensor and data packets collected by the camera sensor in the driverless vehicle.

Data acquisition method and apparatus for driverless vehicle

The present application discloses a data flow processing method and apparatus for a data flow system. A specific implementation of the method includes: acquiring a to-be-processed data flow, and determining, according to a data flow processing instruction, at least one data flow processing node corresponding to the to-be-processed data flow and a passing order in which the to-be-processed data flow passes through the at least one data flow processing node; and connecting together the at least one data flow processing node according to the passing order to obtain a data flow processing channel, and importing the to-be-processed data flow to the data flow processing channel for data processing. This implementation improves the utilization of data flow processing nodes and the data flow processing efficiency.

Data flow processing method and apparatus for data flow system

A method and an apparatus for processing laser point cloud data includes obtaining laser point data to be used by a data receiver comprising an acquisition time; determining a timestamp for representing the acquisition time, and splitting the timestamp into a base timestamp and an offset timestamp; and storing the base timestamp and compressed laser point cloud data. Laser point cloud data output by a laser radar is compressed and comprises only offset timestamps corresponding to respective laser points. The base timestamp and the offset timestamp may be added to obtain the required synchronization precision timestamp, and the data is synchronized. The processing speed of a CPU or GPU for the laser point cloud data is improved while the timestamp precision reaches the precision required by synchronization of the laser point cloud data, and storage space is saved.

Method and apparatus for processing laser point cloud data

A motion compensation method and apparatus applicable to laser point cloud data includes acquiring laser point cloud data of a frame; using an acquisition time of a start point as a target time; interpolating coordinate transformation relations and an end laser point, to obtain coordinate transformation relations of other laser points, and transforming coordinates of the other laser points to the coordinates at the target time. During motion compensation on each frame of laser point cloud data, the coordinate transformation relations corresponding to only the start and end laser points are obtained, and the coordinate transformation relations corresponding to the two points are interpolated to obtain transformation relations corresponding to other laser points, thus performing motion compensation. Therefore, the processing speed is improved and meets high real-time requirements on systems such as autopilot control systems.

Motion compensation method and apparatus applicable to laser point cloud data

The present application discloses a method and an apparatus for processing point cloud data. The method of an embodiment comprises: recognizing an object in a to-be-labeled point cloud frame by using an obstacle recognition algorithm, to obtain a recognition result; presenting the recognition result as an initial labeling result of the point cloud frame; and updating the labeling result in response to a correction operation by a user on the labeling result. According to the embodiment, the speed and accuracy of point cloud data labeling are improved.

Method and apparatus for labeling point cloud data

The present application discloses a data acquiring method and apparatus applied to a driverless vehicle. A specific implementation of the method includes: selecting, from at least one laser radar of the driverless vehicle, a laser radar having a highest importance level as a first laser radar; acquiring a start time of a current time window and executing following data processing steps: executing a real-time acquisition and storage operation on point cloud data packets collected by the each of the at least one laser radar after the start time of the current time window; and determining whether any condition in a group of conditions is satisfied; and setting the start time of the current time window to be the current time and continuing to execute the data processing steps, in response to determining that any of the group of conditions is satisfied. This implementation implements the alignment and storage of point cloud data packets collected by at least one laser radar of the driverless vehicle.

Data acquiring method and apparatus applied to driverless vehicle

The present application discloses a monitoring method and apparatus. A specific implementation of the method comprises: reading information about monitoring data acquisition methods of monitored objects, the monitoring data acquisition method comprising an active acquisition method and a passive acquisition method; executing following steps for each of the monitored objects having the active acquisition method as the monitoring data acquisition method: generating a monitoring data acquisition task of the monitored object; determining a monitoring data acquisition frequency of the monitored object; and adding the monitoring data acquisition task to to-be-executed monitoring data acquisition task sets in a defined period corresponding to the monitoring data acquisition frequency; executing, in successive defined periods, monitoring data acquisition tasks in the corresponding to-be-executed monitoring data acquisition task sets; and parsing acquired monitoring data of the monitored objects to generate a monitoring result. The implementation solves the problem of high system resource occupation and waste of system resources during the monitoring process.

Monitoring method and apparatus

The present application discloses a data acquisition method and apparatus for a driverless vehicle. A specific implementation of the method comprises: selecting a sensor having the highest importance level among at least one sensor of the driverless vehicle as a first sensor and at least one sensor other than the first sensor to forma second sensor set; acquiring a start time of a current time window, and executing the following data processing steps: executing real-time acquiring and storing operations on data packets collected by each sensor among the at least one sensor after the start time of the current time window; determining whether any condition in the following condition group is met; and in response to determining that any condition in the condition group is met, setting the start time of the current time window to the current time, and continuing executing the data processing steps. This implementation realizes synchronized storage of the data packets collected by the at least one sensor of the driverless vehicle.

This disclosure discloses a method and apparatus for transmitting data in a robot operating system. The robot operating system includes a transmitting node, a receiving node, and a shared memory as a transmission medium between the transmitting and receiving nodes. The method in a particular embodiment includes: traversing, by the transmitting node, a sequence of data templates stored in advance in the shared memory, and determining whether the respective data templates in the sequence of data templates are currently being written into or read from; identifying a data template in the sequence of data templates currently being neither written into nor read from as a target data template, and obtaining information of the target data template; and writing data into the target data template according to the information of the target data template. This embodiment can improve the performance of transmitting the data while occupying less memory resources.

Growing community of inventors

Beijing, China

Chengliang Deng