Abstract: A guidance system for remotely guiding a work machine along a virtual path. The system uses a vision system to capture image data representative of areas surrounding the work machine. The image data is used to produce a spatial map. Analysis of image data allows the work machine's then-current position to be represented on the spatial map. A virtual path extending from the work machine's position is next added to the spatial map. The virtual path may be generated in response to external input provided at the display showing an image of the spatial map. Using continuously-updated image data, the work machine is driven toward the virtual path. During operation, the actual path of the work machine is compared to the virtual path. If any deviation between the paths is detected, the trajectory of the work machine is automatically adjusted.

Abstract: Safety metrics and/or statistical system assurance of a particular package of autonomy driving software may be substantially measured using data collected from manually driving a vehicle in the real word, simulations of scenarios which may be faced by a vehicle in the real world, simulations executed with actual software and/or hardware on a vehicle, and/or end-to-end testing of a vehicle in a test environment. Safety metrics and performance of AV software and hardware may be further evaluated through vehicle-in-the-loop testing. During each test scenario, a corresponding set of simulated perception data may be injected to the systems of the autonomous vehicle to cause the autonomous vehicle to react and behave as if one or more simulated objects described by the set of simulated perception data were in the environment of the autonomous vehicle. Each test scenario may be triggered to be performed based on, for example, the autonomous vehicle's location within a real-world vehicle testing space.

Abstract: The invention relates to a motorized flying craft (10) for measuring the contour of a plurality of regions of interest (12a, 12b, 12c) of a surface of a predetermined object (9) to be inspected, said flying craft (10) comprising a carrier frame (20) and motorized means (11, 13) for lifting and moving said carrier frame (20).

Abstract: Surgical systems and methods are disclosed, including, in one embodiment, a system having a coupling system disposed on a distal end portion of a surgical robot arm that secures a navigation array to the robot arm in a plurality of different orientations. The system further includes a navigation system configured to determine a precise location of the distal end portion by measuring a precise location of the navigation array by visually observing the navigation array, receiving a location of the coupling system via one or more encoders in the robot arm, determining the orientation of the navigation array relative to the robot arm based on the visual observation of the navigation array and the received location of the coupling system, and determining the precise location of the distal end portion of the surgical robot arm based on a known spatial relationship between the distal end portion and the coupling system.

Abstract: The object recognition device includes at least one sensor, a data reception unit, and a setting unit. The sensor detects an object present in a detectable range and transmits object data including a plurality of physical quantities indicating a state of the object. The data reception unit receives the object data from the sensor. The setting unit generates prediction data which predicts current object data based on at least a part of past object data received by the data reception unit and a motion model in the object, and sets a association possible region for determining a association between the prediction data and the current object data based on a plurality of physical-quantity delay times each corresponding to each of the plurality of physical quantities included in the prediction data.

Abstract: Systems, devices, and methods for receiving image data; transferring the captured image data to a server having a processor and addressable memory via a network-connected computing device; storing the captured image data on the server; generating captured image metadata based on the stored captured image data; providing access to the captured image data and captured image metadata via an image management component; displaying, by the image management component, the captured image data; and filtering, by the image management component, the captured image data based on the generated captured image metadata.

Abstract: A travel control device is configured to make an autonomous driving vehicle travel in such a way that the autonomous driving vehicle arrives at a specified position specified by a user who intends to board the autonomous driving vehicle, when the autonomous driving vehicle has reached a predetermined range from the specified position, transmit an information sending request notifying the user terminal that the autonomous driving vehicle has reached a vicinity of the specified position and requesting sending of position identifying information for identifying a position at which the user intends to board the autonomous driving vehicle to the user terminal via a communication circuit, and when receiving the position identifying information via the communication circuit, change a position of the autonomous driving vehicle, based on the position identifying information in such a way that the autonomous driving vehicle comes close to the user.

Abstract: The disclosed technology provides solutions for improving passenger drop-off functions implemented by an autonomous vehicle (AV). In some implementations, a process of the disclosed technology can include steps for collecting environmental data about an environment around an autonomous vehicle, wherein the environmental data comprises data pertaining to a roadway navigated by the autonomous vehicle, processing the environmental data to generate an area grid comprising a plurality of grid sections, and associating, based on the environmental data, one or more features with each of the plurality of grid sections. Systems and machine-readable media are also provided.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for generating training data for training a machine learning model to perform trajectory prediction. One of the methods includes: obtaining a training input, the training input including (i) data characterizing an agent in an environment as of a first time and (ii) data characterizing a candidate trajectory of the agent in the environment for a first time period that is after the first time. A long-term label for the candidate trajectory that indicates whether the agent actually followed the candidate trajectory for the first time period is determined. A short-term label for the candidate trajectory that indicates whether the agent intended to follow the candidate trajectory is determined. A ground-truth probability for the candidate trajectory is determined. The training input is associated with the ground-truth probability for the candidate trajectory in the training data.

Abstract: A device for estimating a friction coefficient between a road surface and an automotive tire through determination of a steering torque during a steering maneuver of a slow-rolling or stationary vehicle includes a computer configured for constructing a brush model for a description of the steering torque across a contact patch between the tire and road surface. The steering torque is a torque acting on a steering axis required to overcome resistance to tire twisting on the road surface at a wheel velocity and a steering rate. The steering torque depends on a tire brush vertical load distribution and relative motion of tire brushes and the road surface. The device further includes sensors for measuring the wheel velocity and the steering rate and mechanism for measurements or estimation of the steering torque. The friction coefficient is estimated based on the measurements or estimation of the steering torque and the brush model.

Abstract: The present disclosure relates to methods and systems for generating a channel for use as a spatio-temporal constraint for motion planning for an autonomous vehicle. The method generates a triangulation mesh for a space in an environment. The triangulation mesh includes a plurality of nodes, each represents a geographic location in the space. Some of the nodes correspond to dynamic objects. Based on the triangulation mesh, a candidate channel extending from a starting location to a target location of the autonomous vehicle is generated. Further, the method predicts a time and location of a future triangulation mesh topology event caused by at least one of the nodes that will impact the candidate channel. A valid channel segment for the channel is selected, which extends from the starting location to a channel segment end location preceding the predicted location of the future triangulation mesh topology event. The channel improves the validity of motion planning for a longer time period.

Abstract: A system includes a primary control module, a stability status module, and a supervisory control module. The primary control module is configured to determine at least one control action for at least one of an electronic limited slip differential and an aerodynamic actuator of a vehicle based on a driver command. The stability status module is configured to determine whether at least one component of the vehicle is stable or unstable based on an input from a sensor on the vehicle. The at least one component includes at least one of a vehicle body, a front axle, a rear axle, front wheels, and rear wheels. The supervisory control module is configured to adjust the at least one control action when the at least one component is unstable.

Abstract: Systems and methods for generating semantic occupancy maps are provided. In particular, a computing system can obtain map data for a geographic area and sensor data obtained by the autonomous vehicle. The computer system can identify feature data included in the map data and sensor data. The computer system can, for a respective semantic object type from a plurality of semantic object types, determine, by the computing system and using feature data as input to a respective machine-learned model from a plurality of machine-learned models, one or more occupancy maps for one or more timesteps in the future, and wherein the respective machine-learned model is trained to determine occupancy for the respective semantic object type. The computer system can select a trajectory for the autonomous vehicle based on a plurality of occupancy maps associated with the plurality of semantic object types.

Abstract: A computer includes a processor and a memory, the memory storing instructions executable by the processor to input a signal received from a portable device to a machine learning program trained to output a location of the portable device relative to a vehicle, collect operating data of one or more vehicle components, predict an action of a vehicle user based on the predicted location, and, based on the predicted action of the vehicle user, actuate one or more vehicle components. The machine learning program is trained with a training dataset that is updatable to include the signal, the output predicted location, the collected operating data, the predicted action, and an identified action performed by the vehicle user.

Abstract: Systems and methods for testing a machine learning algorithm or technique of an autonomous driving feature of a vehicle utilize a sensor system configured to capture input data representative of an environment external to the vehicle and a controller configured to receive the input data from the sensor system and perform a testing procedure for the autonomous driving feature that includes inserting known input data into a target portion of the input data to obtain modified input data, processing the modified input data according to the autonomous driving feature to obtain output data, and determining an accuracy of the autonomous driving features based on a comparison between the output data and the known input data.

Abstract: The invention relates to a method for creating a probabilistic free space map with static (2a, 2b, 3) and dynamic objects (V1-V7), having the following steps: retrieving (S1) static objects (2a, 2b, 3) as well as a perception area polygon (WP) from an existing environment model; collecting (S2) predicted trajectories (T1, T2) of dynamic objects (V1-V7); merging (S3) the static objects (2a, 2b, 3) of the perception area polygon (WP) and the predicted trajectories (T1, T2) in a first free space map; fixing (S4) a maximum prediction time; fixing (S5) prediction time steps; fixing (S6) a current prediction time and setting this current prediction time to the value 0 in order to fix the start of a fixed prediction time period; fixing (S7) confidence regions (K) around the static (2a, 2b, 3) and dynamic objects (V1-V7); fixing (S8) at least one uncertain region (U) around at least one static (2a, 2b, 3) or dynamic object (V1-V7); producing (S9) a first probabilistic free space map for the current prediction time

Abstract: A computer-implemented method for localizing a vehicle can include accessing, by a computing system comprising one or more computing devices, a machine-learned retrieval model that has been trained using a ground truth dataset comprising a plurality of pre-localized sensor observations. Each of the plurality of pre-localized sensor observations has a predetermined pose value associated with a previously obtained sensor reading representation. The method also includes obtaining, by the computing system, a current sensor reading representation obtained by one or more sensors located at the vehicle. The method also includes inputting, by the computing system, the current sensor reading representation into the machine-learned retrieval model.

Abstract: A sensor calibration system periodically receives scene data from a detector in a calibration scene. The calibration scene includes calibration targets. The sensor calibration system generates a calibration map based on the scene data. The calibration map is a virtual representation of the calibration scene and includes features of the calibration targets that can be used as ground truth features for calibrating AV sensors. The sensor calibration system can periodically update the calibration map. For instance, the sensor calibration system receives the scene data at a predetermined frequency and updates the calibration map every time it receives new scene data. The predetermined frequency may be a frequency of the detector completing a full scan of the calibration scene. The sensor calibration system provides a latest version of the calibration map for being used by an AV to calibrate a sensor on the AV 110.

Abstract: A system includes a vehicle control module, a vehicle gateway module, and a wired vehicle communications network communicatively coupling the vehicle gateway module to the vehicle control module. The vehicle control module is programmed to receive sensor data from at least one sensor, execute a machine-learning program trained to determine whether the sensor data satisfies at least one criterion, and transmit the sensor data satisfying the at least one criterion to the vehicle gateway module. The vehicle gateway module is programmed to transmit the machine-learning program to the vehicle control module; upon receiving the sensor data from the vehicle control module, store the sensor data; and upon establishing a connection with a remote server, transmit the sensor data to the remote server.

Abstract: A computer-implemented method for using machine learning to determine control parameters for a control system, in particular of a motor vehicle, in particular for controlling a driving operation of the motor vehicle. The method includes: providing a set of travel trajectories; deriving reward functions from the travel trajectories, using an inverse reinforcement learning method; deriving driver type-specific clusters based on the reward functions; determining control parameters for a particular driver type-specific cluster.