Abstract: The present disclosure provides a parking interaction method and apparatus, a storage medium, an electronic device and a vehicle, and belongs to the field of vehicle technologies. The method includes: outputting and displaying a currently-generated virtual parking image and a corresponding real parking image; receiving a parking space selection instruction for the virtual parking image, where the parking space selection instruction includes first position information of a to-be-parked target available parking space in the virtual parking image; according to a correspondence between the virtual parking image and the real parking image, determining second position information of the target available parking space in the real parking image; according to the first position information, highlighting the target available parking space in the virtual parking image, and according to the second position information, highlighting the target available parking space in the real parking image.

Abstract: The present disclosure provides a method and an apparatus for detecting a state of a traffic light, a storage medium a device and a vehicle and belongs to the field of traffic technologies. The method includes: obtaining state detection results for the traffic light based on traffic light images; according to a plurality of state detection results within a first time window and a traffic light state corresponding to each preset road junction type, determining a target road junction type of a target road junction; in response to that a target detection result is present in the state detection results obtained after determining the target road junction type, determining the target detection result as a false detection result, where the target detection result includes a state other than the traffic light state corresponding to the target road junction type.

Abstract: The present disclosure provides an image recognition method, apparatus and device. In the method, after a plurality of to-be-recognized images of a preset target are obtained, first image recognition is performed on the plurality of to-be-recognized images to obtain a plurality of recognition result vectors, and further, the plurality of recognition result vectors are input a preset image quality determination model, where the preset image quality determination model is configured to, based on vector distances between the plurality of recognition result vectors corresponding to the plurality of to-be-recognized images, determine a quality of the plurality of to-be-recognized images. Thus, based on the quality, second image recognition is performed. In this way, the quality of the to-be-recognized images is considered during image recognition, which solves the problems of missing detections and low recognition result accuracy in the existing image recognition.

Abstract: The present invention provides a method and an apparatus for determining travel path bound, a vehicle, a storage medium and a terminal. The method includes: under a first coordinate system, according to contour point positions respectively corresponding to a vehicle and an obstacle in a path trajectory, determining a collision-anticipated overlap corner position; based on a trajectory normal of the vehicle and the overlap corner position, determining a contour corner overlap line and a decision corner reference line; if a first intersection position between the contour corner overlap line and the decision corner reference line is not located on the contour edge, based on the overlap corner position, the decision corner reference line and contour corner positions on the contour corner overlap line, determining a collision-avoiding distance of the vehicle; under a second coordinate system, based on the collision-avoiding distance, generating a travel path bound.

Abstract: The present disclosure provides an image processing method and apparatus, a system, a medium, a device and a driving system, which belong to the field of data processing technologies. The image processing method mainly includes: according to perception target features of each perception target feature class in a plurality of predetermined perception target feature classes, and a pixel requirement of a neural network image perception model corresponding to each perception target feature class in a plurality of predetermined neural network image perception models, performing corresponding adjustment to a current environmental image to obtain a plurality of current environmental feature images; by using the corresponding neural network image perception model, perceiving a perception target in each current environmental feature image to obtain a plurality of perception results; and fusing the plurality of perception results to obtain perception result data of the current environmental image.

Abstract: An embodiment of the present disclosure provides a method and an apparatus for fusing multi-sensor data. The method includes: obtaining a current vision perception result and a current first radar perception result; based on a current vision perception feature corresponding to each current vision perception target in each current vision perception result and a target vision feature fusion model, determining a fusion vision feature corresponding to each current vision perception target; based on the fusion vision feature corresponding to each current vision perception target, a current first radar perception feature corresponding to each current first radar perception target in the current first radar perception result and a pre-constructed radar vision feature fusion model, determining mutually-matched current vision perception targets and current first radar perception targets and corresponding current fusion perception features. In this way, accurate fusion on multi-sensor data can be achieved.

Abstract: The present disclosure provides a method and an apparatus for planning an automatic parking path, a medium and a device and belongs to the field of self driving technologies. The method includes: extracting perception information within a predetermined parking space including a target vehicle place, which is perceived by a vehicle; based on the perception information, identifying a ground sideline of the target parking place, a road sideline and an actual obstacle contour line; cropping the ground sideline and the road sideline intersecting with the ground sideline and taking remaining ground sideline and road sideline as corresponding boundaries of an available parking space of the target parking place; and, performing collision detection based on the corresponding boundaries of the available parking space and the actual obstacle contour line, and based on a result of the collision detection, performing vehicle trajectory planning to obtain an automatic parking planning path.

Abstract: An embodiment of the present disclosure provides a method and an apparatus for constructing a vehicle dynamics model and a method and an apparatus for predicting vehicle state information. The method of constructing a vehicle dynamics model includes: obtaining sample historical state information and a sample control parameter sequence corresponding to each sample time of a target vehicle and label vehicle state information of each sample time; inputting the sample historical state information and the sample control parameter sequence corresponding to the sample time into an initial vehicle dynamics model to determine sample prediction state information; by using the sample prediction state information and the label vehicle state information, determining a current loss value; based on the current loss value, adjusting model parameters of the initial vehicle dynamics model until the initial vehicle dynamics model reaches a preset convergence state so as to obtain a pre-constructed vehicle dynamics model.

Abstract: Provided are a method for generating a lane changing decision-making model and a method and an apparatus for lane changing decision-making of an unmanned vehicle. The method for generating a lane changing decision-making model includes: obtaining a training sample set of vehicular lane changing, wherein the training sample set includes a plurality of training sample groups, each of the training sample groups includes a training sample under each time step length in a process that the vehicle completes lane changing based on a planned lane changing trajectory, the training sample includes a group of state variables and corresponding control variables; obtaining the lane changing decision-making model by training a decision-making model based on deep reinforcement learning network by use of the training sample set, wherein the lane changing decision-making model enables the state variable of the target vehicle and the corresponding control variable to be correlated.