A system and method of detecting and mitigating an erratic vehicle by a host vehicle. The method includes gathering sensor information on a calibratable external region surrounding the host vehicle; analyzing the sensor information to detect a target vehicle traveling in a lane and a movement of the target vehicle in the lane; determining whether the movement of the target vehicle in the lane is erratic; if erratic then designating target vehicle as erratic vehicle; assigning a risk score to the erratic vehicle; and implementing a predetermined mitigating action correlating to the assigned risk score to the erratic vehicle. The mitigating action includes one or more of: warning an operator of the host vehicle, warning a vehicle proximal to the host vehicle, and taking at least partial control of the host vehicle to further distance the host vehicle apart from the erratic vehicle.

Methods and systems are provided for detecting and responding to dangerous road users. In some aspects, a process can include steps for receiving sensor data of a detected object from an autonomous vehicle, determining whether the detected object is exhibiting a dangerous attribute, generating output data based on the determining of whether the detected object is exhibiting the dangerous attribute, and updating a machine learning model based on the output data relating to the dangerous attribute.

Systems and methods are provided for implementing hybrid prediction. Hybrid prediction integrates two deep learning based trajectory prediction approaches: grid-based approaches and graph-based approaches. Hybrid prediction techniques can achieve enhanced performance by combining the grid and graph approaches in a manner that incorporates appropriate inductive biases for different elements of a high-dimensional space. A hybrid prediction framework processor can generate trajectory predictions relating to movement of agents in a surrounding environment based on a prediction model generating using hybrid prediction. Trajectory predictions output from the hybrid prediction framework processor can be used to control an autonomous vehicle. For example, the autonomous vehicle can perform safety-aware and autonomous operations to avoid oncoming objects, based on the trajectory predictions.

Ego actions for a mobile robot in the presence of at least one agent are autonomously planned. In a sampling phase, a goal for an agent is sampled from a set of available goals based on a probabilistic goal distribution, as determined using an observed trajectory of the agent. An agent trajectory is sampled, from a set of possible trajectories associated with the sampled goal, based on a probabilistic trajectory distribution, each trajectory of the set of possible trajectories reaching a location of the associated goal. In a simulation phase, an ego action is selected from a set of available ego actions and based on the selected ego action, the sampled agent trajectory, and a current state of the mobile robot, (i) behaviour of the mobile robot, and (ii) simultaneous behaviour of the agent are simulated, in order to assess the viability of the selected ego action.

A method for monitoring the environment of a vehicle includes evaluating physical measurement data obtained from the environment of the vehicle to determine whether at least one person is approaching the vehicle, how many people approach the vehicle may also be recorded. The method includes evaluating physical measurement data obtained from the environment of the vehicle to determine whether at least one person is moving away from the vehicle and, if appropriate, the number of people that are moving away from the vehicle is also recorded. The method further includes carrying out a check as to whether the number of people that have moved away from the vehicle corresponds to the number of people that have previously approached the vehicle. In response to the check resulting in a difference, it is determined that the vehicle is in an unsafe state.

A method for operating a control device of a motor vehicle driving by automation. The method includes determining a location of the motor vehicle, and acquiring driving-environment data of the motor vehicle, a control characteristic of the control device of the motor vehicle being formed in such a way that a driving behavior of at least one other road user is influenced in defined manner.

According to one aspect, systems and techniques for lane selection may include receiving a current state of an ego vehicle and a traffic participant vehicle, and a goal position, projecting the ego vehicle and the traffic participant vehicle onto a graph network, where nodes of the graph network may be indicative of discretized space within an operating environment, determining a current node for the ego vehicle within the graph network, and determining a subsequent node for the ego vehicle based on identifying adjacent nodes which may be adjacent to the current node, calculating travel times associated with each of the adjacent nodes, calculating step costs associated with each of the adjacent nodes, calculating heuristic costs associated with each of the adjacent nodes, and predicting a position of the traffic participant vehicle.

Systems and methods for countering the usage of autonomous or semi-autonomous vehicles for ramming attacks on a roadway are disclosed. Digital representations of physical trajectories (e.g., roadway travel routes) across which vehicles are expected or permitted to travel are generated based at least on travel-related data (e.g., sensor readings) received from the vehicles over wireless networks. The disclosed systems and methods further generate digital representations of physical trajectories across which vehicles are not permitted to travel, such that the impermissible physical trajectories constitute a deviation from a safe travel route. Additional travel-related data is continuously received from the vehicles in real-time, and the additional data may be combined with non-vehicle data (e.g., pedestrian travel data) and compared to the generated digital representations of permissible and impermissible physical trajectories to determine if the vehicles' physical trajectory is indicative of a harmful impermissible physical trajectory, such as a vehicular ramming attack.

An Autonomous Vehicle (AV) system, including: a tracking subsystem configured to detect and track relative positioning of another vehicle that is behind or lateral to an AV configured to comply with a safety driving model, and to check a safety driving model compliance status of the other vehicle; and a risk reduction subsystem configured to plan, based on the safety driving model compliance status of the other vehicle, an AV action, wherein if the safety driving model compliance status of the other vehicle is unknown or is known to be non-compliant, the AV action is administration of a safety driving model compliance test to the other vehicle, or is a maneuver by the AV to reduce risk of collision with a leading vehicle positioned in front of the AV.

This application relates to the automated driving field, and discloses a method for identifying abnormal driving behavior, a system, and a vehicle including the system. The method for identifying abnormal driving behavior includes: obtaining vehicle driving behavior data, and determining, based on the vehicle driving behavior data, whether a vehicle is in a state of suspicious abnormal driving behavior; obtaining current vehicle driving scenario data if the vehicle is in the state of suspicious abnormal driving behavior; and determining, based on the vehicle driving behavior data and the current vehicle driving scenario data, whether the suspicious abnormal driving behavior is abnormal driving behavior. In the technical solutions of this application, current driving scenario information is introduced to an identification process of abnormal driving behavior of the vehicle, so that accuracy of identifying the abnormal driving behavior is improved.