A method for predicting a trajectory of a target vehicle in an environment of a vehicle. The method includes the steps of a) capturing states of the target vehicle, capturing states of further vehicle objects in the environment of the vehicle and capturing road markings by a camera-based capture device; b) preprocessing the data obtained in step a), wherein outliers are removed and missing states are calculated; c) calculating an estimated trajectory by a physical model on the basis of the data preprocessed in step b); d) calculating a driver-behavior-based trajectory on the basis of the data preprocessed in step b); and e) combining the trajectories calculated in steps c) and d) to form a predicted trajectory of the target vehicle.

Techniques for performing multi-sensor calibration on a vehicle are described. A method includes obtaining, from each of at least two sensors located on a vehicle, sensor data item of a road comprising a lane marker, extracting, from each sensor data item, a location information of the lane marker, and calculating extrinsic parameters of the at least two sensors based on determining a difference between the location information of the lane marker from each sensor data item and a previously stored location information of the lane marker.

A route selection device includes a processor configured to identify a position of a lane on which a vehicle is traveling; search for candidate partial routes leading from a current position of the vehicle to a waypoint on a route leading from a start point to a destination, the waypoint being located between the current position and the destination; determine a lane change location where a lane change will be made for each of the candidate partial routes found by searching; and select, as a partial route, a candidate partial route having a minimum total score regarding the lane change location from the candidate partial routes found by searching. The score is weighted depending on the position of the determined lane change location or whether a lane change at the lane change location can be controlled by a travel controller.

In various examples, a multi-sensor fusion machine learning model—such as a deep neural network (DNN)—may be deployed to fuse data from a plurality of individual machine learning models. As such, the multi-sensor fusion network may use outputs from a plurality of machine learning models as input to generate a fused output that represents data from fields of view or sensory fields of each of the sensors supplying the machine learning models, while accounting for learned associations between boundary or overlap regions of the various fields of view of the source sensors. In this way, the fused output may be less likely to include duplicate, inaccurate, or noisy data with respect to objects or features in the environment, as the fusion network may be trained to account for multiple instances of a same object appearing in different input representations.

Embodiments of the present disclosure disclose a method for position detection, a device, and a storage medium. The method includes: detecting a first lane line in a current image captured by a camera; performing an optimization on an initial transformation matrix reflecting a mapping relationship between a world coordinate system and a camera coordinate system based on a detection result of the first lane line; and obtaining a first 3D coordinate of a target object in the current image according to a transformation matrix optimized, and determining an ultimate 3D coordinate of the target object according to the first 3D coordinate.

A road shape recognizer includes a peripheral information recognizer that recognizes at least two items of peripheral information based on an output of a periphery detector. A reliability assigner assigns a reliability level to each of the peripheral information. A point sequence generator generates and places a point sequence representing a shape of a road on which the own vehicle travels, based on at least two items of peripheral information and the reliability level. The point sequence generator generates and places a point sequence by generating and placing points one by one toward a distant place from a point located at a prescribed relative position to the own vehicle. The point sequence generator generates and places the next point corresponding to an amount of change in shape and a position of a point generated and placed at the end of the point sequence. The amount of change in shape is represented by the peripheral information and determined per section having a prescribed distance.

A driving information display apparatus and method corrects camera pose values using a vanishing point. The apparatus includes a processor configured to receive driving guidance and vehicle location information, control the output of a guidance screen, and set a crop area in an image from a forward-view camera based on the calculated vanishing point. The storage unit stores road information, algorithms, and camera pose values, allowing the processor to generate a guidance screen by integrating the driving guidance information within the crop area centered on the vanishing point.

Target point cloud data about a specific road are extracted from perimeter point cloud data acquired by moving a road surface measurement device along a measurement route and scanning the surroundings thereof. A data storage unit stores trajectory point sequence data that represent, as a plurality of trajectory points, the perimeter point cloud data and a trajectory of the movement of the road surface measurement device. A trajectory point sequence setting unit acquires a trajectory point sequence at equal intervals from the trajectory point sequence data. An extraction area setting unit sets, as extraction areas, a column area Ci and a parallelepiped area Hi that are geometric areas disposed at predetermined positions below a trajectory point Xi. An approximate nearest neighbor search processing unit and an extraction processing unit extract, as the target point cloud data, point data that belong to this extraction area of the perimeter point cloud data.

Vehicle-mounted camera pose estimation methods, apparatuses, and systems, and electronic devices involve performing lane line detection of a road on which a vehicle drives on the basis of a video stream of the road acquired by a vehicle-mounted camera; obtaining horizon information of the road on which the vehicle drives according to a lane line detection result; and obtaining pose information of the vehicle-mounted camera according to the horizon information.

Methods, systems, and media for determining characteristics of roads are provided. In some embodiments, the method comprises: receiving, at a first time point, first camera information from a camera associated with a vehicle; identifying a first position of a feature of an object in front of the vehicle based on the first camera information; receiving, at an additional time point, additional camera information from the camera; identifying an updated position of the feature of the object in front of the vehicle based on the additional camera information; determining a relative motion of the feature of the object in front of the vehicle based on the first position and the updated position; and determining a characteristic of a road the vehicle is on based on the relative motion of the feature of the object in front of the vehicle.