Methods, systems, and apparatus, including computer programs encoded on computer storage media, for predicting near-curb driving behavior. One of the methods includes obtaining agent trajectory data for an agent in an environment, the agent trajectory data comprising a current location and current values for a predetermined set of motion parameters of the agent; processing a model input generated from the agent trajectory data using a trained machine learning model to generate a model output comprising a prediction of whether the agent will exhibit near-curb driving behavior within a predetermined timeframe, wherein an agent exhibits near-curb driving behavior when the agent operates within a particular distance of an edge of a road in the environment; and using the prediction to generate a planned path for a vehicle in the environment.

A driving assistance device to properly recognize a relative positional relationship between mobile objects without the use of map information. The driving assistance device includes a relative position judgment section for judging a relative positional relationship of an object of interest and a first neighboring object relative to a second neighboring object, based on object-of-interest information and neighbor information, to make a judgment as a first judgment on a relative positional relationship between the object of interest and the first neighboring object, based on the aforementioned judgment result.

Location polygons are defined along traffic lanes and parking spaces to facilitate determination of the location of a vehicle relative to features associated with the location polygons. The location polygons are used, in one application, to identity entrance and exit of a special toll lane along a roadway, and to ensure that the vehicle properly enters and exits the tolling lane. The location polygons define geofenced regions, and each definition for a geofenced region can include one or more rules that are used to evaluate location information reported by a user’s equipment. The rules dictate whether an action it taken or inhibited, such as charging a toll or not charging a toll, based on other location information reported by the user’s equipment.

The present disclosure provides a vehicle travel control method and apparatus. A specific implementation lies in: acquiring a distance between a vehicle and a first intersection, where the first intersection is an intersection for the vehicle to go across on a first road which the vehicle is currently on; acquiring, on a determination that the distance is less than or equal to a preset distance, intersection information of the first intersection and travelling information of the vehicle, where the intersection information includes lane information of at least two lanes on the first road; determining, according to the intersection information and the travelling information, a target weight of each of the lanes for the vehicle to travel into; and determining, according to the target weight of each of the lanes for the vehicle to travel into, an intended travel route of the vehicle.

An automated vehicle control distributed network node, that includes at least two modems for communicating with two neighboring roadside nodes on the same side of the roadway; at least one antenna for communicating with vehicles via a wireless connection; pattern recognition processing operative to detect patterns using image data from a plurality of high speed, high resolution video cameras that include night vision; vehicle prediction processing, operatively coupled to the pattern recognition processing, operative to predict vehicle location, velocity and direction using the pattern recognition processing; and a vehicle controller, operatively coupled to the vehicle prediction processing to receive vehicle prediction data, and to the at least one antenna, operative to send acceleration, deceleration and steering control signals to a plurality of vehicles in response to vehicle prediction data received from the vehicle prediction processing.

Technologies and techniques for laterally controlling a motor vehicle on a road, where a roadway marking assigned to a driver's side of the motor vehicle is assigned to a lane type based on sensor data. The lane type characterizes whether the roadway marking is assigned to an ego lane on which the motor vehicle is to be guided, or whether it is assigned to a neighboring lane adjacent to the ego lane. The roadway marking is also assigned to the lane type, based on swarm data, and the lane type assigned to the roadway marking is established as the lane type that is determined as a function of the swarm data when the sensor lane type and the swarm data lane type diverge. The motor vehicle is laterally controlled as a function of the established lane type of the roadway marking assigned to the driver's side of the motor vehicle.

A method for monitoring uncertainty for human-like behavioral modulation of trajectory planning includes: retrieving map and agent information of a current driving state of an autonomously operated host automobile vehicle; dividing uncertainty conditions affecting a trajectory of the host automobile vehicle into an expected uncertainty and an unexpected uncertainty; calculating the expected uncertainty in a first operation branch by forming attention zones according to identified portions of lanes which may potentially collide with a planned route of the host automobile vehicle; determining the unexpected uncertainty in a second operation branch by calculating an anomaly score for any other vehicles in a surrounding area of the host automobile vehicle positioned in the lanes which may potentially collide with the planned route of the host automobile vehicle; and modulating trajectory operation signals determined for the expected uncertainty if the unexpected uncertainty meets or exceeds a predetermined threshold.

An apparatus and a method in an autonomous vehicle detect a traffic flow obstruction target. The apparatus detects information about at least one of a speed of another vehicle, a driving path of the other vehicle, or a position of the other vehicle. The apparatus calculates a degree to which the other vehicle interferes with traffic flow, based on the detected information and based on high definition map information stored in a memory. The apparatus selects a traffic flow obstruction target, based on the degree to which the other vehicle interferes with the traffic flow. The apparatus detects a target causing bypass driving, which is present on a driving path, to enhance the continued operation of autonomous driving.

According to one aspect, routing of an autonomous or semi-autonomous vehicle may be based at least in part upon the availability of particular geographical areas through which the vehicle may be routed. A geographical area may be defined, based on constraints such as road conditions, lane features, operational constraints, etc., such that at least one routing path between each of one or more sources and one or more destinations within the geographical area may be traversed by a vehicle. Constraints may be temporary and may be accounted for substantially in real-time when defining and/or updating a geographical area within which a vehicle is to operate. In some instances, an inclusion constraint can be defined based on a frequency of use of a given routing path associated with one or more points of interest for a geographical area. Accounting for constraints enables autonomous vehicles to be deployed safely and efficiently.

This document describes methods and systems for enabling an autonomous vehicle (AV) to determine a path to a stopping location. The AV will determine a desired stop location (DSL) that is associated with a service request. The AV's motion control system will move the AV along a path to the DSL. While moving along the path, the AV's perception system will detect ambient conditions near the DSL. The ambient conditions will be parameters associated with a stopping rule. The AV will apply the stopping rule to the ambient conditions to determine whether the stopping rule permits the AV to stop at the DSL. If the stopping rule permits the AV to stop at the DSL, the motion control system will move the AV to, and stop at, the DSL. Otherwise, the motion control system will not stop the AV at the DSL.