Systems and methods for situational modification of autonomous vehicle operation are disclosed. According to aspects, a computing device may detect the occurrence of an emergency event and may determine a current operation of an autonomous vehicle that may be associated with the emergency event. The computing device may determine a modification to operation of the autonomous vehicle, where the modification may represent a violation of a roadway regulation that may enable effective handling of the emergency event. The computing device may generate a set of instructions for the autonomous vehicle to execute to cause the autonomous vehicle to undertake the operation modification.

A computer implemented method for self-learning of a control system. The method includes creating an initial knowledge base. The method learns first principles using the knowledge base. The method creates initial control commands derived from the knowledge base. The method generates constraints for the control commands. The method performs constrained reinforcement learning by executing the control commands with the constraints and observing feedback to improve the control commands. The method enriches the knowledge base based on the feedback.

There is provided a system for adapting parameters of a vehicle for reduction of likelihood of an adverse event, comprising: hardware processor(s) executing a code for: performing, for each respective driver of multiple drivers: obtaining an indication of a vehicle driven by the respective driver, obtaining an indication of a certain advanced driver assistance system (ADAS) selected from multiple ADAS for installation in the vehicle, obtaining an environmental profile indicative of a prediction of an environment in which the vehicle with installed ADAS is predicted for driving therein at a future time interval, defining a simulation model in which the vehicle with installed ADAS is driving according to the environment profile, computing a risk of an adverse event during the future time interval by executing the simulation model, and selecting parameter(s) of the vehicle for adaptation thereof according to a predicted likelihood of reducing the risk of the adverse event.

An embodiment driving pattern learning apparatus includes a driving mode selection unit configured to receive an autonomous driving mode or a driving pattern learning mode using an input unit, a driving pattern learning unit configured to learn a driving pattern of a user comprising acceleration, braking, steering, an inter-vehicle distance, a lane change, overtaking, or a response to road facilities based on a process of driving by the user in the driving pattern learning mode, and a memory configured to store the driving pattern learned by the driving pattern learning unit.

Discussed is a method of predicting an intelligent beam of an autonomous vehicle in an autonomous driving system. The method can include obtaining sensing information for detecting one or more adjacent objects through at least one sensor of the autonomous vehicle, in response to an occurrence of a blockage event where a blocker detected on a line of sight (LOS) path between the autonomous vehicle and a target vehicle blocks the target vehicle, selecting some of a plurality of non-line of sight (NLOS) paths between the autonomous vehicle and the target vehicle to continue communication between the autonomous vehicle and the target vehicle, and selecting an optimal beam related to the target vehicle based on the selected one or more of the plurality of NLOS paths. The selecting of some of the plurality of NLOS paths can be performed based on a pre-trained machine learning network

A traveling assistance apparatus controls a safety apparatus for avoiding collision between an own vehicle and an object, based on detection information from an object detection apparatus. The assistance apparatus calculates a predicted time to collision of a target object and the own vehicle, and operates the safety apparatus in response to the predicted time to collision being equal to or less than a predetermined operation timing. In response to a steering operation by a driver for collision avoidance of the own vehicle being performed, the assistance apparatus performs operation-stop in which operation of the safety apparatus is stopped or operation-delay in which the operation timing is delayed. The assistance apparatus suppresses the operation-stop or the operation-delay in response to a target object serving as a collision avoidance target being recognized in a path of the own vehicle after the steering operation by the driver, based on the detection information.

A controller receives sensor data during a ride and provides it to a server system. A passenger further provides feedback concerning the ride in the form of some or all of an overall rating, flagging of ride anomalies, and flagging of road anomalies. The sensor data and feedback are input to a training algorithm, such as a deep reinforcement learning algorithm, which updates an artificial intelligence (AI) model. The updated model is then propagated to controllers of one or more autonomous vehicle which then perform autonomous navigation and collision avoidance using the updated AI model.

A system and method for limited range sharing of vehicle setup information is disclosed. The system includes a memory; and at least one processor configured to: obtain a vehicle setup information for a vehicle, the vehicle setup information comprising one or more adjustment settings for one or more adjustable components of the vehicle; package the vehicle setup information into at least one tune; format the at least one tune in a computer readable scannable format; and present the at least one tune in the computer readable scannable format.

The present disclosure relates to a navigation system for a host vehicle. The system may include a processing device programmed to receive, from a camera, a plurality of images representative of an environment of the host vehicle; analyze the plurality of images to identify at least one pedestrian in the environment of the host vehicle; cause at least one adjustment of a navigational system of the host vehicle to signal to the pedestrian a navigational intent of the host vehicle; analyze the plurality of images to detect a potential reaction of the pedestrian to the at least one adjustment of the navigational system of the host vehicle; determine a navigational action for the host vehicle based on a detected potential reaction of the pedestrian; and cause at least one adjustment of a navigational actuator of the host vehicle in response to the determined navigational action for the host vehicle.

A computer implemented method for self-learning of a control system. The method includes creating an initial knowledge base. The method learns first principles using the knowledge base. The method creates initial control commands derived from the knowledge base. The method generates constraints for the control commands. The method performs constrained reinforcement learning by executing the control commands with the constraints and observing feedback to improve the control commands. The method enriches the knowledge base based on the feedback.