An autonomous driving system acquires information concerning a vehicle density in an adjacent lane that is adjacent to a lane on which an own vehicle is traveling, when the own vehicle travels on a road having a plurality of lanes. The autonomous driving system selects the adjacent lane as an own vehicle travel lane, when the vehicle density in the adjacent lane that is calculated from the acquired information is lower than a threshold density that is determined in accordance with relations between the own vehicle and surrounding vehicles. The autonomous driving system performs lane change to the adjacent lane autonomously, or propose lane change to the adjacent lane to a driver, when the adjacent lane is selected as the own vehicle travel lane.

An autonomous driving control device is capable of starting an autonomous driving control without an operation of a driver and reducing a possibility that the driver can not start manual driving. An autonomous driving control is switched to manual driving when a determination section determines that the amount of operation by the driver is equal to or greater than a first threshold, before a predetermined time elapses since the autonomous driving control is automatically started. An autonomous driving control is switched to a manual driving when the determination section determines that the amount of operation by the driver is equal to or greater than a second threshold that is greater than the first threshold, after the predetermined time elapses.

Systems, methods and apparatus may be configured to implement automatic semantic classification of a detected object(s) disposed in a region of an environment external to an autonomous vehicle. The automatic semantic classification may include analyzing over a time period, patterns in a predicted behavior of the detected object(s) to infer a semantic classification of the detected object(s). Analysis may include processing of sensor data from the autonomous vehicle to generate heat maps indicative of a location of the detected object(s) in the region during the time period. Probabilistic statistical analysis may be applied to the sensor data to determine a confidence level in the inferred semantic classification. The inferred semantic classification may be applied to the detected object(s) when the confidence level exceeds a predetermined threshold value (e.g., greater than 50%).

A method and a system determine visibility distances based on a dynamic Field of View (FoV) of a subject vehicle. A vehicle incorporates the system. Map polygons are created, each of which determines edges of a road in a map of the surroundings of the subject vehicle. Further, visible areas are determined in the map by intersecting the map polygons with the dynamic FoV. Based on the visible areas, a visibility distance for the road is determined.

A drive assistant (700) executes a drive assist function that is a function of a drive assist system. A first electronic control apparatus (401) has a first sensor (501). A second electronic control apparatus (402) has a second sensor (502). The first electronic control apparatus (401) is connected to the drive assistant (700) via a main network (10). The second electronic control apparatus (402) is connected to the first electronic control apparatus (401) via a sub-network (20) having no connection to the drive assistant (700). The first electronic control apparatus (401) outputs, to the main network (10), control assist information generated on the basis of first sensing information acquired by the first sensor (501) and second sensing information acquired by the second sensor (502).

A method for determining event data including: sampling a first data stream within a first time window at a first sensor of an onboard vehicle system coupled to a vehicle, extracting interior activity data from the first data stream; determining an interior event based on the interior activity data; sampling a second data stream within a second time window at a second sensor of the onboard vehicle system; extracting exterior activity data from the second image stream; determining an exterior event based on the exterior activity data; correlating the exterior event and the interior event to generate combined event data; automatically classifying the combined event data to generate an event label; and automatically labeling the first time window of the first data stream and the second time window of the second data stream with the combined event label to generate labeled event data.

A vehicle control apparatus, a vehicle control method, and a program that can curb unnecessary driving control are provided. The vehicle control apparatus includes a pedestrian recognition unit configured to recognize a crossing pedestrian crossing a road on which a vehicle travels, a space recognition unit configured to recognize whether there is a space having a predetermined width or more between a lane on which the vehicle travels and an oncoming lane, and a driving control unit configured to execute avoidance support for avoiding contact between the vehicle and the crossing pedestrian recognized by the pedestrian recognition unit based on a behavior of the crossing pedestrian and a behavior of the vehicle, in which the driving control unit is configured to determine whether the crossing pedestrian recognized by the pedestrian recognition unit is moving from the oncoming lane side to a space recognized by the space recognition unit, and curb the avoidance support upon determination that the crossing pedestrian is moving to the space.

A vehicle control device includes a recognition unit which recognizes surrounding situations of a vehicle, and a driving control unit which automatically controls at least steering of the vehicle based on the surrounding situations recognized by the recognition unit, and the driving control unit increases a distance between the vehicle and a traffic participant in the case where the recognition unit recognizes the traffic participant as an overtaking target and a predetermined structure in a traveling direction of the vehicle, and the vehicle travels on a side opposite to the predetermined structure in a road width direction in a case that viewed from the traffic participant to overtake the traffic participant, as compared to a case where the traffic participant as the overtaking target is recognized by the recognition unit in the traveling direction of the vehicle and the predetermined structure is not recognized.

In one embodiment, a method is provided. The method includes receiving, at an input of a first machine learning model, first input data representing an environment. The method also includes determining, by the first machine learning model, a set of objects within the environment based on the first input data. The method further includes determining, by a second machine learning model, a set of behaviors for a second set of objects. An input of the second machine learning model is coupled to a set of intermediate layers of the first machine learning model. Determining the set of objects and determining the set of behaviors for the second set of objects is performed at least partially simultaneously.

Systems and methods for operating a vehicle are disclosed. The methods comprise: generating, by a computing device, a vehicle trajectory for the vehicle while the vehicle is in motion; detecting an object within a given distance from the vehicle; generating at least one possible object trajectory for the object which was detected; performing a collision check to determine whether a collision between the vehicle and the object can be avoided based on the vehicle trajectory and the at least one possible object trajectory; performing a plausibility check to determine whether the collision is plausible based on content of a map, when a determination is made in the collision check that a collision between the vehicle and the object cannot be avoided; and performing operations to selectively cause the vehicle to perform an emergency maneuver based on results of the plausibility check.