According to an aspect of an embodiment, operations may comprise receiving a plurality of frame sets generated while navigating a local environment, receiving an occupancy map (OMap) representation of the local environment, for each of the plurality of frame sets, generating, using the OMap representation, one or more instances each comprising a spatial cluster of neighborhood 3D points generated from a 3D sensor scan of the local environment, and classifying each of the instances as dynamic or static, tracking instances classified as dynamic across the plurality of frame sets using a tracking algorithm, assigning a single instance ID to tracked instances classified as dynamic across the plurality of frame sets, estimating a bounding box for each of the instances in each of the plurality of frame sets, and employing the instances as ground truth data in a training of one or more deep learning classifiers.

A computer-implemented method of adaptively controlling an autonomous operation of a vehicle is provided. The method includes steps of (a) in a critic network in a computing system configured to autonomously control the vehicle, determining, using samples of passively collected data and a state cost, an estimated average cost, and an approximated cost-to-go function that produces a minimum value for a cost-to-go of the vehicle when applied by an actor network; and (b) in an actor network in the computing system and operatively coupled to the critic network, determining a control input to apply to the vehicle that produces the minimum value for the cost-to-go, wherein the actor network is configured to determine the control input by estimating a noise level using the average cost, a cost-to-go determined from the approximated cost-to-go function, a control dynamics for a current state of the vehicle, and the passively collected data.

A computer-implemented method of adaptively controlling an autonomous operation of a vehicle is provided. The method includes steps of (a) in a critic network in a computing system configured to autonomously control the vehicle, determining, using samples of passively collected data and a state cost, an estimated average cost, and an approximated cost-to-go function that produces a minimum value for a cost-to-go of the vehicle when applied by an actor network; and (b) in an actor network in the computing system and operatively coupled to the critic network, determining a control input to apply to the vehicle that produces the minimum value for the cost-to-go, wherein the actor network is configured to determine the control input by estimating a noise level using the average cost, a cost-to-go determined from the approximated cost-to-go function, a control dynamics for a current state of the vehicle, and the passively collected data.

Systems and methods for determining object prioritization and predicting future object locations for an autonomous vehicle are provided. A method can include obtaining, by a computing system comprising one or more processors, state data descriptive of at least a current or past state of a plurality of objects that are perceived by an autonomous vehicle. The method can further include determining, by the computing system, a priority classification for each object in the plurality of objects based at least in part on the respective state data for each object. The method can further include determining, by the computing system, an order at which the computing system determines a predicted future state for each object based at least in part on the priority classification for each object and determining, by the computing system, the predicted future state for each object based at least in part on the determined order.

The present disclosure is generally related to a data processing system to validate vehicular functions in a voice activated computer network environment. The data processing system can improve the efficiency of the network by discarding action data structures and requests that invalid prior to their transmission across the network. The system can invalidate requests by comparing attributes of a vehicular state to attributes of a request state.

A technique for enabling efficient retrieval of a digital representation of personality data of a user (402) by a client device (406) from a server (404) is disclosed, wherein the digital representation of the personality data is processed at the client device (406) to provide a user-adapted service to the user (402). A method implementation of the technique is performed by the server (404) and comprises storing a neural network being trained to compute personality data of a user based on input obtained from the user (402), receiving, from the client device (406), a request for a digital representation of personality data for a user (402), and sending, to the client device (406), the requested digital representation of the personality data of the user (402), wherein the personality data of the user is computed using the neural network based on input obtained from the user (402).

A technique for enabling efficient retrieval of a digital representation of personality data of a user (402) by a client device (406) from a server (404) is disclosed, wherein the digital representation of the personality data is processed at the client device (406) to provide a user-adapted service to the user (402). A method implementation of the technique is performed by the server (404) and comprises storing a neural network being trained to compute personality data of a user based on input obtained from the user (402), receiving, from the client device (406), a request for a digital representation of personality data for a user (402), and sending, to the client device (406), the requested digital representation of the personality data of the user (402), wherein the personality data of the user is computed using the neural network based on input obtained from the user (402).

A method is for training an object detector configured to detect objects in sensor data of a sensor. The method includes providing first sensor data of the sensor, providing an object representation assigned to the first sensor data, and transmitting the object representation to a sensor model. The method further includes imaging object representations onto the first sensor data of the sensor with the sensor model, assigning the object representation to second sensor data with the sensor model, and training the object detector based on the second sensor data.

A method is for training an object detector configured to detect objects in sensor data of a sensor. The method includes providing first sensor data of the sensor, providing an object representation assigned to the first sensor data, and transmitting the object representation to a sensor model. The method further includes imaging object representations onto the first sensor data of the sensor with the sensor model, assigning the object representation to second sensor data with the sensor model, and training the object detector based on the second sensor data.

A method of maintaining stability of a motor vehicle having a first axle, a second axle, and a steering actuator configured to steer the first axle includes determining localization and heading of the vehicle. The method also includes determining a current side-slip angle of the second axle and setting a maximum side-slip angle of the second axle using the friction coefficient at the vehicle and road surface interface. The method additionally includes predicting when the maximum side-slip angle would be exceeded using the localization, heading, and determined current side-slip angle as inputs to a linear computational model. The method also includes updating the model using the prediction of when the maximum side-slip angle would be exceeded to determine impending instability of the vehicle. Furthermore, the method includes correcting for the impending instability using the updated model and the maximum side-slip angle via modifying a steering angle of the first axle.