Abstract: A system and method for contextually monitoring vehicle states. The method includes enriching a plurality of messages based on at least one missing property value and a plurality of predictive properties in the plurality of messages, wherein at least one message of the plurality of messages is generated by a vehicle monitoring system, wherein each message of the plurality of messages is ingested at a vehicle state monitor, wherein each of the predictive properties is a property indicated in one of the plurality of messages of a predetermined type of property; and generating a contextual vehicle state for a vehicle based on the enriched plurality of messages, wherein the contextual vehicle state is a snapshot of a plurality of vehicle state properties at a given time.

Abstract: In one embodiment, a first trajectory is generated for a driving environment using control values allowed for a driving-by-wire system. If the trajectory includes a collision with an object, the ADV estimates the time of the collision and the relative speed between the ADV and the object at the time of the collision. A second trajectory is then generated for the driving environment using control values allowed for a human driver. The time of the collision and the relative speed between the ADV and the object at the time collision on the second trajectory are also estimated. The ADV then compares the two collision times and the two relative speeds, and based on the comparison, generates an alert message for the human driver to take over the control of the ADV.

Abstract: Disclosed are a robot and a method for delivering articles by the robot. A robot, according to an embodiment of the present disclosure, includes: one or more loading spaces configured to load articles; a travel driver configured to move the robot; an input interface configured to obtain identification information of the articles; a measurer configured to measure temperature of each of the articles; a temperature regulator configured to heat or cool the loading spaces; and a controller configured to determine a type of each of the articles from the identification information, determine a delivery route based on the determined type of article, and control the temperature regulator based on the measured temperature while controlling the travel driver such that the robot moves along the determined delivery route. Examples of the present disclosure are implemented by executing a machine learning algorithm in a 5G environment connected for Internet of things (IoT).

Abstract: A method for checking a capacity of at least one supply line for an electrically operated assembly that is coupled electrically via the at least one supply line to an electrical system of an at least partially automated mobile platform. The method includes: ensuring a non-critical operating state of the mobile platform; determining an off-load terminal voltage at an input connection of the assembly, with zero-current supply line; applying a defined current to the supply line of the electrically operated assembly; determining an on-load terminal voltage at the input connection of the assembly, in doing so, the defined voltage being applied to the supply line of the electrically operated assembly; determining a differential voltage between the off-load terminal voltage and the on-load terminal voltage; comparing the differential voltage to a differential-voltage limit value to determine whether the capacity of the electric supply line is sufficient to operate the assembly.

Abstract: A vehicle agnostic removable pod can be mounted on a vehicle using one or more legs of a pod mount. The removable pod can collect and time stamp a variety of environmental data as well as vehicle data. For example, environmental data can be collected using a sensor suite which can include an IMU, 3D positioning sensor, one or more cameras, and/or a LIDAR unit. As another example, vehicle data can be collected via a CAN bus attached to the vehicle. Environmental data and/or vehicle data can be time stamped and transmitted to a remote server for further processing by a computing device.

Abstract: A computer includes a processor and a memory storing instructions executable by the processor to identify a virtual boundary between a host roadway lane of a host vehicle and a target roadway lane of a target vehicle, the virtual boundary based on a predicted path of the target vehicle, determine a first constraint value based on a boundary approach velocity of the target vehicle, determine a second constraint value based on (1) a boundary approach velocity of the host vehicle and (2) a boundary approach acceleration of the host vehicle and perform a threat assessment of a collision between the host vehicle and the target vehicle upon determining that the first constraint value violates a first threshold or the second constraint value violates a second threshold.

Abstract: In a vehicle system that can receive remote support from a remote support server (e.g., interfacing with a human or computer teleoperator), a local normalization engine locally normalizes operation of the vehicle based on locally available sensor data that may not be accessible to the remote support server. The local normalization engine applies transformations to control commands received from the remote support server to transform the command to compensate for conditions that are locally sensed and may be unknown to the remote support server. Alternatively, or in addition, the local normalization engine controls auxiliary functions of the vehicle (e.g., by activating one or more auxiliary actuators) that may not be under direct control of the remote support server.

Abstract: A control system for a work vehicle includes an actual topography acquisition device, a storage device, a soil amount acquisition device, and a controller. The actual topography acquisition device acquires actual topography information indicating an actual topography of a work target. The storage device stores design topography information indicating a final design topography. The soil amount acquisition device generates a soil amount signal indicating a held soil amount of the work implement. The controller acquires the actual topography information from the actual topography acquisition device and acquires the design topography information from the storage device. The controller generates a command signal for moving the work implement at position that is between the actual topography and the final design topography and is a predetermined distance above the actual topography. The controller acquires the soil amount signal and changes the predetermined distance based on the held soil amount.

Abstract: A method includes receiving a component history for a processor module, the component history including identities of host vehicles in a set of host vehicles in which the processor module was installed; receiving an operational history for the set of host vehicles for a time period the processor module was installed on the set of host vehicles; receiving indirect sensor measurements related to the set of host vehicles for the time period; receiving a part survival model that is based at least in part on a part status of a plurality of historical processor modules, the plurality of historical processor modules having a historical component history, a historical operational history, and historical indirect sensor measurements; and determining a survival probability of the processor module based at least in part on the component history, the operational history, the indirect sensor measurements, and the part survival model.

Abstract: Multi-agent reinforcement learning may be used for rider order-dispatching via matching the distribution of orders and vehicles. Information may be obtained. The information may include a plurality of vehicle locations of a plurality of vehicles, a plurality of ride orders, and a current time. The obtained information may be input into a trained model. The trained model may be based on Kullback-Leibler divergence optimization and independent agents under a guidance of a joint policy. A plurality of order-dispatching tasks may be generated for the plurality of vehicles to fulfill.

Abstract: A managing apparatus for positioning rental vehicles includes a memory storing instructions and a processor configured to execute the instructions to cause the managing apparatus to access model information and location information for a plurality of autonomous vehicles, receive a request including a delivery location and a chosen model for renting, select an autonomous vehicle of the chosen model from among the plurality of autonomous vehicles based on the model information, the location information, and the delivery location, instruct the selected autonomous vehicle to fully-autonomously or semi-autonomously travel to the delivery location, and instruct the selected autonomous vehicle to switch to manual operation mode at the delivery location for manual operation by a vehicle rental customer.

Abstract: A vehicle includes a plurality of controllers. During a key-off mode, a master controller may maintain operation of a predetermined set of controllers by broadcasting a power sustain request on a communication bus. Responsive to a loss of communication for a time exceeding a predetermined amount of time with one or more of the predetermined set of controllers, the master controller may be programmed to stop broadcasting the power sustain request to cause the predetermined set of controllers to power down.

Abstract: A vehicle control system is provided that includes a control unit that can be disposed onboard a vehicle to control movement of the vehicle; and one or more transceiver devices that can emit plural signals from the vehicle, with at least one of the plural signals containing a vehicle identifier, and, responsive to a receiver unit disposed off-board the vehicle receiving at least one of the plural signals, the control unit can determine a location of the vehicle and can communicate a signal to the one or more transceiver devices based on the location. The control unit can change the movement of the vehicle based on the vehicle location information.

Abstract: The present disclosure relates to a pull up controlling system of a vehicle for controlling the vehicle in a planned stop-and-go situation. The pull up controlling system determines a position of the vehicle and identifies a stop-and-go destination within a predeterminable distance from the vehicle position. The stop-and-go destination includes an interaction interface. The pull up controlling system provides a user input request relating to whether to discard or acknowledge the stop-and-go destination. The pull up controlling system receives intention data indicative of confirmation to acknowledge the stop-and-go destination. The pull up controlling system further derives preference data indicative of an openable section of the vehicle. The pull up controlling system maneuvers the vehicle to pull up at the stop-and-go destination, with the openable section aligned with the interaction interface.

Abstract: Techniques for using a detector coupled to a controller area network (CAN) bus of a vehicle system to identify and generate alerts in response to wake-up attacks on the vehicle system. Techniques include an electronic control unit (ECU) sending a wake-up message across the CAN bus that is detected by the detector. The detector includes memory and a processor to identify the timestamp of the wake-up message when the vehicle ignition is off. The detector determines a total operational time for the ECU over an observation time period and generates a notification of a wake-up anomaly when the total operational time over the observation time period exceeds a predetermined threshold.

Abstract: A method and apparatus for controlling a vehicle are provided. A lane in which a target vehicle is driving, and an object in a vicinity of the lane are detected from an image of surroundings of the target vehicle, a degree of danger of the object is evaluated, driving information of the target vehicle is determined based on the degree of danger, and the target vehicle is controlled based on the driving information.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for training a neural network having a plurality of sub neural networks to assign respective confidence scores to one or more candidate future trajectories for an agent. Each confidence score indicates a predicted likelihood that the agent will move along the corresponding candidate future trajectory in the future. In one aspect, a method includes using the first sub neural network to generate a training intermediate representation; using the second sub neural network to generate respective training confidence scores; using a trajectory generation neural network to generate a training trajectory generation output; computing a first loss and a second loss; and determining an update to the current values of the parameters of the first and second sub neural networks.

Abstract: A charging control system includes a lawn mower that has a battery and performs a lawn mowing work while traveling autonomously, and a charging station for charging the battery. The lawn mower includes a period calculator for calculating a shutoff period of supply power supplied from the charging station, and a first communication unit. The charging station includes a second communication unit communicating with the first communication unit, an information acquisition unit for acquiring shutoff period information indicating the shutoff period from the first communication unit via the second communication unit, a switch for shutting off the supply power, and a shutoff controller for controlling the operation of the switch. The shutoff controller releases the shutoff of the power supply to the lawn mower based on the shutoff period information.

Abstract: The technology disclosed provides a so-called “pointer sentinel mixture architecture” for neural network sequence models that has the ability to either reproduce a token from a recent context or produce a token from a predefined vocabulary. In one implementation, a pointer sentinel-LSTM architecture achieves state of the art language modeling performance of 70.9 perplexity on the Penn Treebank dataset, while using far fewer parameters than a standard softmax LSTM.

Abstract: A system and method to distribute pesticides, fertilizers, water, and other materials on a farm with accuracy and precision is disclosed in order to combat the problems imposed on the environment due to over-fertilization and over use of pesticides. This system and method is a social networking control system in which multiple farms have independent grids of sensors capable of detecting the presence of pesticides, fertilizers, water, and other materials in the air, in the top-soil, and in the groundwater. These grids of sensors detect the location and concentration of these materials and reports them back to a social control system for analysis. The control system regulates the deposition of further chemicals through computer control of the chemical dispersal systems.