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

Beijing Baidu Netcom Science and Technology Co., Ltd.

Baidu USA LLC

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

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

Baidu Online Network Technology Co., Ltd.

Ruigang Yang

Xinjing Cheng

Feixiang Lu

Hao Xu

Yajue Yang

Liang Wang

Yu Ma

Jin Fang

Feilong Yan

Peng Wang

Tongtong Zhao

Chi Zhang

Xibin Song

Sibo Zhang

Dingfu Zhou

He Jiang

Miao Liao

Feihu Zhang

Binbin Cao

Zhibin Hong

Wei Xu

Jinxin Zhao

Yanfu Zhang

Chenye Guan

A method, device and a computer-readable storage medium for optimizing simulation data are provided. The method for optimizing simulation data can include: inputting the simulation data generated by a simulator to a first generative adversarial network comprising a migration model; and optimizing the simulation data generated by the simulator with the migration model to generate optimized simulation data. In an embodiment of the present application, the simulation data is optimized by the generative adversarial network to enable the simulation data closer to the real data in representation. Therefore, the quality and accuracy of the simulation data can be ensured, the validity and reliability of the simulation data can be improved at least to some extent, and the cost for constructing the simulator can also be reduced.

Method and device for optimizing simulation data, and computer-readable storage medium

A method for generating simulated point cloud data, a device, and a storage medium includes: acquiring at least one frame of point cloud data collected by a road collecting device in an actual environment without a dynamic obstacle as static scene point cloud data; setting, least one dynamic obstacle in a coordinate system matching the static scene point cloud data; simulating in the coordinate system, a plurality of simulated scanning lights emitted by a virtual scanner located at an origin of the coordinate system; updating the static scene point cloud data according to intersections of the plurality of simulated scanning lights and the at least one dynamic obstacle to obtain the simulated point cloud data comprising point cloud data of the dynamic obstacle; and at least one of adding a set noise to the simulated point cloud data, and, deleting point cloud data corresponding to the dynamic obstacle according to a set ratio.

Method for generating simulated point cloud data, device, and storage medium

A device for detecting a rotary angle used for an excavator. The device can include: a synchronous belt, arranged around a rotating shaft of a slewing mechanism of the excavator, a tooth-shaped surface of the synchronous belt being away from a surface of the rotating shaft; a transmission part, engaged with the tooth-shaped surface of the synchronous belt and arranged on a supporting base; an angle detection part, in transmission connection to the transmission part; and the supporting base, connected to a chassis of the excavator.

Device for detecting rotary angle and excavator

Embodiments of the present disclosure relate to a method and apparatus for controlling an excavator to excavate. The method includes: acquiring a two-dimensional image of a material pile; generating a three-dimensional model of the material pile based on the two-dimensional image; analyzing the three-dimensional model to determine a target excavating point and a target excavating trajectory of the material pile; and controlling an excavator to excavate a material at the target excavating point along the target excavating trajectory.

Method and apparatus for controlling excavator to excavate

Embodiments of the present disclosure relate to a three-dimensional reconstruction method and apparatus for a material pile, an electronic device, and a computer-readable medium. The method may include: acquiring, in response to an instruction for controlling an excavator body of an excavator to rotate to transport materials being detected, a sequence of depth images of an excavated material pile collected by a binocular camera provided on a side of the excavator; and performing three-dimensional reconstruction based on the sequence of depth images of the material pile, to generate a three-dimensional model of the material pile.

Three-dimensional reconstruction method and apparatus for material pile, electronic device, and computer-readable medium

Presented herein are novel embodiments for converting a given speech audio or text into a photo-realistic speaking video of a person with synchronized, realistic, and expressive body dynamics. In one or more embodiments, 3D skeleton movements are generated from the audio sequence using a recurrent neural network, and an output video is synthesized via a conditional generative adversarial network. To make movements realistic and expressive, the knowledge of an articulated 3D human skeleton and a learned dictionary of personal speech iconic gestures may be embedded into the generation process in both learning and testing pipelines. The former prevents the generation of unreasonable body distortion, while the later helps the model quickly learn meaningful body movement with a few videos. To produce photo-realistic and high-resolution video with motion details, a part-attention mechanism is inserted in the conditional GAN, where each detailed part is automatically zoomed in to have their own discriminators.

Personalized speech-to-video with three-dimensional (3D) skeleton regularization and expressive body poses

Embodiments of the present disclosure disclose an apparatus and method for fixing a sensing system to an unmanned construction machinery. The fixing apparatus includes an internal fixing module and an external fixing module, in which, the internal fixing module is configured to fix a sensing device in a sensing system of an unmanned construction machinery to the external fixing module in a predetermined fixing manner; and the external fixing module is configured to control the sensing device in the sensing system to perform an environment sensing on a target area of the unmanned construction machinery in a predetermined control manner.

Fixing apparatus and method of sensing system

A method and an apparatus for controlling an excavator are provided according to the embodiments. The method includes: determining, according to preset value ranges of trajectory parameters in at least two trajectory parameters, trajectory parameter value combinations, to obtain a first trajectory parameter value combination set; determining, from the first trajectory parameter value combination set, trajectory parameter value combinations for moving from a preset starting position to a preset end position, to obtain a second trajectory parameter value combination set; determining, in the second trajectory parameter value combination set, a trajectory parameter value combination that satisfies a preset excavation condition, to obtain a third trajectory parameter value combination set; determining a target excavation trajectory from the third trajectory parameter value combination set; and sending a control instruction to the excavator to enable the excavator to excavate material according to the target excavation trajectory.

Method and apparatus for controlling excavator

Embodiments of the present disclosure relate to a method and apparatus for controlling vehicle driving. The method includes: generating a global path of a driving site of a vehicle; executing following controlling: generating a local path of a site in front of a current position of the vehicle based on the global path, the local path following a direction of the global path, controlling the vehicle to drive along the local path until reaching an end point of the local path, determining whether the vehicle reaches an end point of the global path, and terminating the controlling if the vehicle reaches the endpoint of the global path; and continuing, in response to determining the vehicle failing to reach the end point of the global path, executing the controlling.

Method and apparatus for controlling vehicle driving

Embodiments of the present disclosure relate to a method and apparatus for outputting information. The method includes: acquiring a target excavating trajectory, the target excavating trajectory including at least two sub-trajectories; determining trajectory parameters of the at least two sub-trajectories; determining, based on the trajectory parameters, positions of a plurality of control points; and outputting the positions of the plurality of control points.

Method and apparatus for outputting information

Presented are systems and methods for improving speed and quality of real-time per-pixel depth estimation of scene layouts from a single image by using an end-to-end Convolutional Spatial Propagation Network (CSPN). An efficient linear propagation model performs propagation using a recurrent convolutional operation. The affinity among neighboring pixels may be learned through a deep convolutional neural network (CNN). The CSPN may be applied to two depth estimation tasks, given a single image: (1) to refine the depth output of existing methods, and (2) to convert sparse depth samples to a dense depth map, e.g., by embedding the depth samples within the propagation procedure. The conversion ensures that the sparse input depth values are preserved in the final depth map and runs in real-time and is, thus, well suited for robotics and autonomous driving applications, where sparse but accurate depth measurements, e.g., from LiDAR, can be fused with image data.

Systems and methods for depth estimation via affinity learned with convolutional spatial propagation networks

Embodiments of the present disclosure disclose a control system and a control method of an unmanned engineering machinery. The system includes at least a slave computer, a master computer and an execution device. The slave computer is configured to receive a current sensing value fed back by a sensing device in the unmanned engineering machinery, and to send the current sensing value to the master computer. The master computer is configured to generate a control instruction according to the current sensing value and a predetermined target sensing value of the sensing device, and to send the control instruction to the execution device through the slave computer.

Control system and control method of unmanned engineering machinery

A method and system for simulating a distribution of obstacles are provided. The method includes: acquiring a plurality of point clouds of a plurality of frames, wherein each point cloud includes a plurality of original obstacles; acquiring real labeling data of an acquisition vehicle, and obtaining data of a simulation position of the acquisition vehicle based on the real labeling data and a movement rule of the acquisition vehicle; determining the number of obstacles to be simulated based on the data of the simulation position of the acquisition vehicle; selecting the determined number of obstacles to be simulated, from a range with the simulation position of the acquisition vehicle as a center, wherein the range is less than or equal to a maximum scanning range of the vehicle; and acquiring real labeling data of the selected obstacles, and obtaining a position distribution of the selected obstacles.

Method and device for simulating a distribution of obstacles

Embodiments of the present disclosure relate to a method and apparatus for updating information. The method may include: acquiring road network structure information of a target road network and vehicle information of a target number of vehicles in the target road network, the vehicle information including initial state information, perception information and positioning information, and the vehicle information being constrained by the road network structure information; selecting a target vehicle from the target number of vehicles; determining, based on a vehicle dynamics model, a reference speed at which the target vehicle passes a preset time step; and updating vehicle information of a vehicle in the target road network based on the reference speed of the target vehicle.

Method and apparatus for updating information

Embodiments of an obstacle distribution simulation method, device and terminal based on a probability graph are provided. The method can include: acquiring a plurality of point clouds of a plurality of frames; acquiring real labeling data of an acquisition vehicle at vehicle labeled positions, and acquiring data of a simulation position of the acquisition vehicle; determining the number of obstacles to be simulated at a position to be simulated; extracting real labeling data of the obstacles, and constructing a labeling data set; dividing the labeling data set into a plurality of grids and calculating occurrence probabilities of the plurality of obstacles; selecting the determined number of obstacles to be simulated according to the occurrence probabilities; and acquiring a position distribution of the selected obstacles to be simulated for the position to be simulated based on the real labeling data of the selected obstacles to be simulated.

Growing community of inventors

Beijing, China

Ruigang Yang