The present disclosure relates to an autonomous vehicle, a station system, and a door control method for the autonomous vehicle. An exemplary embodiment of the present disclosure provides an autonomous vehicle, comprising a processor configured to control opening and closing of a door of the autonomous vehicle depending on existence of an object around the door of the autonomous vehicle and whether an object outside and inside a station reaches a boarding zone of the autonomous vehicle within a predetermined time when the autonomous vehicle is stopped, and a storage configured to store data and algorithms driven by the processor.

A computing system can be configured to input data that describes sensor data into an object detection model and receive, as an output of the object detection model, object detection data describing features of the plurality of the actors relative to the autonomous vehicle. The computing system can generate an input sequence that describes the object detection data. The computing system can analyze the input sequence using an interaction model to produce, as an output of the interaction model, an attention embedding with respect to the plurality of actors. The computing system can be configured to input the attention embedding into a recurrent model and determine respective trajectories for the plurality of actors based on motion forecast data received as an output of the recurrent model.

A control system and a method for predicting a location of dynamic objects, for example, of pedestrians, which are able to be detected by the sensors of a vehicle. The control system includes a multitude of sensors and a processing system, which is configured to combine with a first program the objects that are detected by the multitude of sensors to form an object list, each entry of the object list encompassing the location, a speed and an open route for each of the objects, and the object list including a time stamp; and to determine with a second program for at least a portion of the dynamic objects an additional object list from a predefined number of object lists, the additional object list including a time stamp for a future point in time and encompassing at least the location of the dynamic objects.

A method for determining a lane change, performed by an apparatus for determining a lane change of an object located around a driving vehicle with which is equipped a sensor, the method including, detecting a plurality of objects located around the driving vehicle using scanning information obtained repeatedly at every predetermined period of time by the sensor scanning surroundings of the driving vehicle, selecting at least one candidate object estimated to change lanes among the plurality of objects based on previously detected lane edge information and determining whether the candidate object changes lanes based on information on movement of the candidate object.

The present invention relates to a method of cooperatively coordinating future driving maneuvers of a vehicle with fellow maneuvers of at least one fellow vehicle, wherein trajectories for the vehicle are rated with an effort value each, trajectories and fellow trajectories of the fellow vehicle are combined into tuples, the trajectory and the associated effort value of a collision-free tuple are selected as reference trajectory and reference effort value, trajectories with a lower effort value than the reference effort value are classified as demand trajectories, trajectories with higher effort value than the reference effort value are classified as alternative trajectories, and a data packet having a trajectory set consisting of the reference trajectory and the associated reference effort value as well as at least one trajectory from a group comprising the demand trajectories and the alternative trajectories as well as the respective effort values is transmitted to the fellow vehicle.

Among other things, techniques are described for predicting how an agent (e.g., a vehicle, bicycle, pedestrian, etc.) will move in an environment based on prior movement, the road network, the surrounding objects and/or other relevant environmental factors. One trajectory prediction technique involves generating a probability map for an agent's movement. Another trajectory prediction technique involves generating a trajectory lattice, for an agent's movement. In addition, a different trajectory prediction technique involves multi-modal regression where a classifier (e.g., a neural network) is trained to classify the probability of a number of (learned) modes such that each model produces a trajectory based on the current input.

Provided herein is a system of a vehicle that comprises one or more sensors, one or more processors, and memory storing instructions that, when executed by the one or more processors, causes the system to perform: selecting a trajectory along a route of the vehicle; predicting a trajectory of another object along the route; adjusting the selected trajectory based on a predicted change, in response to adjusting the selected trajectory, to the predicted trajectory of the another object, the predicted change to the predicted trajectory of the another object being stored in a model; determining an actual change, in response to adjusting the selected trajectory, to a trajectory of the another object, in response to an interaction between the vehicle and the another object; updating the model based on the determined actual change to the trajectory of the another object; and selecting a future trajectory based on the updated model.

Techniques for accurately predicting and avoiding collisions with objects detected in an environment of a vehicle are discussed herein. A vehicle computing device can implement a model to output data indicating costs for potential intersection points between the object and the vehicle in the future. The model may employ a control policy and a time-step integrator to determine whether an object may intersect with the vehicle, in which case the techniques may include predicting vehicle actions by the vehicle computing device to control the vehicle.

Aspects and implementations of the present disclosure relate to modeling of positional uncertainty of moving objects using precomputed polygons, for example, for the purposes of computing autonomous vehicle (AV) trajectories. An example method includes: receiving, by a data processing system of an AV, data descriptive of an agent state of an object; generating a polygon representative of the agent state; identifying extreme vertices of the polygon along a longitudinal axis parallel to a heading direction of the object or along a lateral axis orthogonal to the heading direction; and applying, based on the extreme vertices, at least one expansion transformation to the polygon along the longitudinal axis or the lateral axis to generate a precomputed polygon.

An autonomous vehicle uses machine learning based models to predict hidden context attributes associated with traffic entities. The system uses the hidden context to predict behavior of people near a vehicle in a way that more closely resembles how human drivers would judge the behavior. The system determines an activation threshold value for a braking system of the autonomous vehicle based on the hidden context. The system modifies a world model based on the hidden context predicted by the machine learning based model. The autonomous vehicle is safely navigated, such that the vehicle stays at least a threshold distance away from traffic entities.