Abstract: A method for training a migration scene-based trajectory prediction model is provided, a first trajectory prediction model and a plurality of candidate training samples are obtained; for any candidate training sample, a reference value corresponding to the candidate training sample is determined according to at least one of a trajectory feature corresponding to the candidate training sample or a prediction result of the first trajectory prediction model for the candidate training sample; target training samples are selected from the plurality of candidate training samples according to the reference values corresponding to the plurality of candidate training samples; and the first trajectory prediction model is trained according to the target training samples, to obtain a second trajectory prediction model, where the second trajectory prediction model is configured to predict traveling trajectories of obstacles in a migration scene.

Abstract: Vehicle control is provided, including: obtaining vehicle information of a target vehicle and environmental information of a reference environment in which the target vehicle is located; obtaining a target matrix based on the vehicle information and the environmental information; splitting the target matrix to obtain a plurality of sub-matrices; and obtaining target driving control information of the target vehicle based on matrix elements in the sub-matrices and driving control information of a surrounding vehicle of the target vehicle.

Abstract: An obstacle trajectory prediction method and apparatus are provided. In some embodiments, a global interaction feature under joint action of a vehicle and obstacles is determined according to historical status information and current status information of the vehicle, historical status information and current status information of the obstacles, and a future motion trajectory planned by the vehicle; an individual interaction feature of a to-be-predicted obstacle is determined according to the global interaction feature and current status information of the to-be-predicted obstacle; and a future motion trajectory of the to-be-predicted obstacle is predicted through the individual interaction feature and information about an environment around the vehicle.

Abstract: A trajectory prediction method and apparatus, a storage medium, and an electronic device are provided. In embodiments of this disclosure, according to a historical trajectory of a designated target and a historical trajectory of each obstacle, a historical interaction feature between the designated target and each obstacle is determined, and a motion trajectory of the designated target is predicted to obtain an initial predicted trajectory. A future interaction feature between each obstacle and the designated target is then determined according to the initial predicted trajectory and a planned trajectory of each obstacle. According to the future interaction feature, a final predicted trajectory of the designated target is obtained.

Abstract: An unmanned device acquires sensing data of surrounding obstacles; determines, for each obstacle, at least one predicted track of the obstacle in a future period of time based on the sensing data; determines, for each moment in the future period of time and according to the predicted track corresponding to the obstacle, a collision probability that a collision with the obstacle occurs at each position in a target region at the moment; and determines a global collision probability that the collision with the obstacle occurs in the entire target region at the moment. According to the global collision probability corresponding to each obstacle at each moment, the unmanned device controls the unmanned device in the future period of time.

Abstract: A method for trajectory planning, an apparatus, a storage medium, and an electronic device are provided. A constraint set of a space including a target device is determined according to a velocity of an unmanned device and velocities of designated obstacles, so that during optimization of a preliminary reference trajectory, a solution can be obtained with the space in the constraint set as a solution space under the constraint of the constraint set, so as to ensure that the solution space is a convex space, and relatively satisfactory reference trajectory points can be solved.

Abstract: Various embodiments a method and apparatus for unmanned driving behavior decision-making and model training, and an electronic device. The method includes: acquiring sample data, wherein the sample data includes a sample image; extracting a sample feature vector corresponding to the sample data, wherein a feature vector of the sample image is extracted by manifold dimension reduction; and based on the sample feature vector, training by semi-supervised learning to obtain a target decision-making model, wherein the target decision-making model is used for decision-making classification.

Abstract: This specification discloses an obstacle trajectory prediction method and apparatus. In embodiments of the present disclosure, a global interaction feature under joint action of a vehicle and obstacles is determined according to historical status information and current status information of the vehicle, historical status information and current status information of the obstacles, and a future motion trajectory planned by the vehicle; an individual interaction feature of a to-be-predicted obstacle is determined according to the global interaction feature and current status information of the to-be-predicted obstacle; and a future motion trajectory of the to-be-predicted obstacle is predicted through the individual interaction feature and information about an environment around the vehicle.

Abstract: An unmanned device acquires sensing data of surrounding obstacles; determines, for each obstacle, at least one predicted track of the obstacle in a future period of time based on the sensing data; determines, for each moment in the future period of time and according to the predicted track corresponding to the obstacle, a collision probability that a collision with the obstacle occurs at each position in a target region at the moment; and determines a global collision probability that the collision with the obstacle occurs in the entire target region at the moment. According to the global collision probability corresponding to each obstacle at each moment, the unmanned device controls the unmanned device in the future period of time.

Abstract: A method for trajectory planning, an apparatus, a storage medium, and an electronic device are provided. A constraint set of a space including a target device is determined according to a velocity of an unmanned device and velocities of designated obstacles, so that during optimization of a preliminary reference trajectory, a solution can be obtained with the space in the constraint set as a solution space under the constraint of the constraint set, so as to ensure that the solution space is a convex space, and relatively satisfactory reference trajectory points can be solved.

Abstract: Embodiments provide extracting information respectively corresponding to predetermined dimensions from environment information corresponding to an unmanned driving environment. In some embodiments, the information respectively corresponding to the dimensions is input into an identification model to obtain a driving feature. Then a risk value representing a driving risk degree of an unmanned device is determined, and a maximum variation of the information corresponding to at least one dimension is determined when a variation of the driving feature is less than a predetermined threshold. A maximum variation of the information corresponding to each dimension is used as a risk contribution feature. A variation representative value of the information corresponding to each dimension is determined from the risk contribution feature. According to the variation representative values of the dimensions, a driving risk factor corresponding to the risk value is determined based on the driving feature.

Abstract: A method and apparatus for controlling an unmanned driving device are provided. An unmanned driving device obtains a travel trajectory of the unmanned driving device as a first trajectory, obtains a travel trajectory of a target object around as a second trajectory, and determines a spatial relation between the unmanned driving device and the target object according to the first trajectory and the second trajectory. Subsequently, the unmanned driving device can code the spatial relation according to a preset coding manner to obtain spatial coding information, where each code in the spatial coding information is configured to represent a spatial relation between the unmanned driving device and the target object at a moment corresponding to the code.

Abstract: A method and apparatus for controlling an unmanned device, a storage medium, and an electronic device are provided. In some embodiments, not only a control policy is determined according to a single frame of a current image, but a safety representation value of an unmanned device at a current moment and safety representation values of the unmanned device at respective historical moments are respectively determined according to a current image and environment images of several historical moments. In those embodiments, a control policy of the unmanned device at a next moment is determined according to a safety representation value of the unmanned device at current moment and the safety representation values of the unmanned device at respective historical moments.

Abstract: A trajectory prediction method and apparatus, a storage medium, and an electronic device are provided. In embodiments of this disclosure, according to a historical trajectory of a designated target and a historical trajectory of each obstacle, a historical interaction feature between the designated target and each obstacle is determined, and a motion trajectory of the designated target is predicted to obtain an initial predicted trajectory. A future interaction feature between each obstacle and the designated target is then determined according to the initial predicted trajectory and a planned trajectory of each obstacle. According to the future interaction feature, a final predicted trajectory of the designated target is obtained.

Abstract: The disclosure discloses a method and device for determining parameter for a gaze tracking device. The method includes that: an image to be detected and a characteristic matrix of the image to be detected are acquired; the image to be detected is processed according to a preset model to obtain a first vector and second vector of the image to be detected; and a parameter of the image to be detected is determined according to the first vector and the second vector, the parameter of the image to be detected including at least one of: a position of a glint in the image to be detected and a gaze direction of eyes in the image to be detected.

Abstract: Vehicle control is provided, including: obtaining vehicle information of a target vehicle and environmental information of a reference environment in which the target vehicle is located; obtaining a target matrix based on the vehicle information and the environmental information; splitting the target matrix to obtain a plurality of sub-matrices; and obtaining target driving control information of the target vehicle based on matrix elements in the sub-matrices and driving control information of a surrounding vehicle of the target vehicle.

Abstract: This application provides a method and apparatus for unmanned driving behavior decision-making and model training, and an electronic device. The method includes: acquiring sample data, wherein the sample data includes a sample image; extracting a sample feature vector corresponding to the sample data, wherein a feature vector of the sample image is extracted by manifold dimension reduction; and based on the sample feature vector, training by semi-supervised learning to obtain a target decision-making model, wherein the target decision-making model is used for decision-making classification.

Abstract: The present disclosure discloses a method and apparatus for filtering glints. The method includes that: a first image and a reference image matched with the first image are acquired, and first-type glints are displayed in the first image; second-type glints in the first image are determined according to the reference image, and the second-type glints are glints obtained by estimating glints in the first image on the basis of the reference image; a matching result of the first-type glints and the second-type glints is determined according to first positions of the first-type glints and second positions of the second-type glints; and the first-type glints are filtered according to the matching result.

Abstract: Embodiments of the present disclosure provide a model generation method, including: constructing a training sample set including a sample image, where feature information of the sample image is line information and optical flow information; and learning the training sample set to generate a recognition model that uses line information and optical flow information of an image as input.

Abstract: The present disclosure discloses a method and apparatus for filtering glints. The method includes that: a first image and a reference image matched with the first image are acquired, and first-type glints are displayed in the first image; second-type glints in the first image are determined according to the reference image, and the second-type glints are glints obtained by estimating glints in the first image on the basis of the reference image; a matching result of the first-type glints and the second-type glints is determined according to first positions of the first-type glints and second positions of the second-type glints; and the first-type glints are filtered according to the matching result.