The present disclosure relates to a method for predicting a behavior of a target vehicle in a maneuvering space, in which a plurality of maneuvers which can be executed in the future are provided for the target vehicle, in the maneuvering space, the target vehicle located in the maneuvering space being observed by a plurality of observation vehicles, for each maneuver of the target vehicle that can be executed in the future, a single probability distribution for the relevant maneuver that can be executed in the future being determined, on the basis of an observation performed by a relevant observation vehicle, for each maneuver of the target vehicle that can be executed in the future, a total probability distribution for the relevant maneuver of the target vehicle that can be executed in the future being established from a plurality of single probability distributions.

The present disclosure relates to a method for predicting a behavior of a target vehicle in a maneuvering space, in which a plurality of maneuvers which can be executed in the future are provided for the target vehicle, in the maneuvering space, the target vehicle located in the maneuvering space being observed by a plurality of observation vehicles, for each maneuver of the target vehicle that can be executed in the future, a single probability distribution for the relevant maneuver that can be executed in the future being determined, on the basis of an observation performed by a relevant observation vehicle, for each maneuver of the target vehicle that can be executed in the future, a total probability distribution for the relevant maneuver of the target vehicle that can be executed in the future being established from a plurality of single probability distributions.

Mobile wireless terminals, such as vehicles in traffic, can determine the relative positions of other vehicles with improved precision by arranging to acquire GNSS (global navigational satellite system) signals simultaneously, and then analyzing the various data sets differentially. Simultaneous acquisition can cancel many important errors such as motional errors of the vehicles, atmospheric distortions, and satellite timebase errors. Differential analysis to determine the relative positions of vehicles (as opposed to their overall geographical coordinates) can reduce errors related to satellite ephemeris and velocity, as well as roundoff errors. Localization with a precision of less than 1 meter can greatly improve collision avoidance while discriminating near-miss scenarios from imminent collisions, according to some embodiments. Messaging examples, in 5G and 6G, to manage the simultaneous acquisition and differential analysis, are provided in examples. Many other aspects are disclosed.

Mobile wireless terminals, such as vehicles in traffic, can determine the relative positions of other vehicles with improved precision by arranging to acquire GNSS (global navigational satellite system) signals simultaneously, and then analyzing the various data sets differentially. Simultaneous acquisition can cancel many important errors such as motional errors of the vehicles, atmospheric distortions, and satellite timebase errors. Differential analysis to determine the relative positions of vehicles (as opposed to their overall geographical coordinates) can reduce errors related to satellite ephemeris and velocity, as well as roundoff errors. Localization with a precision of less than 1 meter can greatly improve collision avoidance while discriminating near-miss scenarios from imminent collisions, according to some embodiments. Messaging examples, in 5G and 6G, to manage the simultaneous acquisition and differential analysis, are provided in examples. Many other aspects are disclosed.

An apparatus, method and computer program product are provided for predicting wrong-way-driving events. A WWD event defines an event in which a vehicle travels in a direction opposing a traffic direction designated for a portion of a road. In one example, the apparatus estimates a likelihood of portion of the road inducing a wrong-way-driving (WWD) event by using a machine learning model as a function of map data, sensor data, or a combination thereof. If the likelihood satisfies a threshold, the apparatus updates a map layer to indicate the WWD event at the portion of the road.

An apparatus, method and computer program product are provided for predicting wrong-way-driving events. A WWD event defines an event in which a vehicle travels in a direction opposing a traffic direction designated for a portion of a road. In one example, the apparatus estimates a likelihood of portion of the road inducing a wrong-way-driving (WWD) event by using a machine learning model as a function of map data, sensor data, or a combination thereof. If the likelihood satisfies a threshold, the apparatus updates a map layer to indicate the WWD event at the portion of the road.

A warning system and method are provided. The warning system includes a plurality of sensing apparatuses and a server. The sensing apparatuses are used for sensing a driving trajectory of each of a plurality of two-wheeled vehicles. The server compares the driving trajectories with an accident hotspot list to determine whether at least one first driving trajectory matches an accident hotspot location, wherein the accident hotspot list is generated by a plurality of driving behavior events corresponding to each of the two-wheeled vehicles. The server generates a warning message to remind a first driver of a first two-wheeled vehicle corresponding to the at least one first driving trajectory when determining that the at least one first driving trajectory matches the accident hotspot location.

A warning system and method are provided. The warning system includes a plurality of sensing apparatuses and a server. The sensing apparatuses are used for sensing a driving trajectory of each of a plurality of two-wheeled vehicles. The server compares the driving trajectories with an accident hotspot list to determine whether at least one first driving trajectory matches an accident hotspot location, wherein the accident hotspot list is generated by a plurality of driving behavior events corresponding to each of the two-wheeled vehicles. The server generates a warning message to remind a first driver of a first two-wheeled vehicle corresponding to the at least one first driving trajectory when determining that the at least one first driving trajectory matches the accident hotspot location.

A system receives confirmation that a vehicle has accepted automatic control imposition for a drive within a geo-fenced boundary. The system tracks travel of a plurality of vehicles, including the vehicle, within the geo-fenced boundary. The system may determine that the vehicle has a threshold likelihood of encountering at least one of another vehicle or a boundary of the geo-fence at a threshold speed or above and responsive to the determination, impose automatic control on the vehicle, including at least one of controlled braking or speed limiting.

A system receives confirmation that a vehicle has accepted automatic control imposition for a drive within a geo-fenced boundary. The system tracks travel of a plurality of vehicles, including the vehicle, within the geo-fenced boundary. The system may determine that the vehicle has a threshold likelihood of encountering at least one of another vehicle or a boundary of the geo-fence at a threshold speed or above and responsive to the determination, impose automatic control on the vehicle, including at least one of controlled braking or speed limiting.