Abstract: The present invention relates to the efficient use of both local and remote computational resources and communication bandwidth to provide distributed environment mapping using a plurality of mobile sensor-equipped devices. According to a first aspect, there is provided a method of determining a global position of one or more landmarks on a global map, the method comprising the steps of determining one or more differences between sequential sensor data captured by one or more moving devices; determining one or more relative localisation landmark positions with respect to the one or more moving devices; determining relative device poses based one or more differences between sequential sensor data relative to the one or more relative localisation landmark positions; and determining a correlation between each device pose and the one or more relative localisation landmarks positions.

Abstract: A vehicle external environment recognition apparatus to be applied to a vehicle includes a road surface determination processor and a three-dimensional object determination processor. The road surface determination processor determines a road surface region that corresponds to a road surface in an image, plots representative distances of respective horizontal lines in the road surface region at respective vertical positions of the horizontal lines, and generates first and second road surface models. The second road surface model represents a farther portion of the road surface region from the vehicle than the first road surface model and differs in a gradient from the first road surface model. On a condition that an angle formed by the first and second road surface models is greater than a predetermined angle, the three-dimensional object determination processor cancels the second road surface model and extends far the first road surface model.

Abstract: Systems, vehicles and methods for determining wrong direction driving are disclosed. In one embodiment, a system for determining a vehicle traveling in a wrong direction includes one or more sensors that produce sensor data, one or more processors, and one or more non-transitory computer-readable medium storing computer readable-instructions. When the computer-readable instructions are executed by the one or more processors, the computer-readable instructions cause the one or more processors to determine one or more lanes within a roadway using the sensor data, determine a direction of travel of the one or more lanes using the sensor data, and identify a non-compliant vehicle traveling in a direction in the one or more lanes that is different from the determined direction of travel in the one or more lanes.

Abstract: A vehicle control method executed by a processor capable of making a subject vehicle change a lane includes: acquiring surrounding information of the subject vehicle by a sensor provided in the subject vehicle; determining whether a distractive factor for a driver of another vehicle is present when the subject vehicle enters the front of the other vehicle traveling on the second lane for changing lanes from a first lane to a second lane adjacent to the first lane; setting a lane change time required for the subject vehicle to change lanes longer than when determining none of the distractive factor to be present when determining distractive the factor to be present; and controlling a traveling position of the subject vehicle on the first lane within the lane change time.

Abstract: A method and system for inspecting and monitoring an elevator installation includes sending an autonomous flying object having at least one sensor to the elevator installation, and granting access to a hoistway of the elevator installation to the autonomous flying object. The autonomous flying object is positioned within the hoistway, and data collected by the associated sensor is sent to a remote elevator service center. The autonomous flying object and the associated sensor can be used to monitor and inspect the elevator installation on a temporary basis, for example for a specific number of hours, days or weeks. Once the autonomous flying object has gained access to the hoistway, the elevator installation can resume normal operation thereby keeping downtime to a minimum. After completing its tasks at one elevator installation, the autonomous flying object can be directed by the remote elevator service center to monitor and inspect another elevator installation.

Abstract: A method for adjusting a safety distance between a first vehicle and a second vehicle preceding the first vehicle at a bunched traffic site by at least one control device is disclosed which includes receiving measurement data of at least one sensor unit with the at least one control device. The method includes evaluating the measurement data of the at least one sensor unit with the at least one control device to identify a bunched traffic site in front of the second vehicle and identifying a safety distance of the first vehicle from the second vehicle, wherein the identified safety distance accounts for a possible reverse motion of the second vehicle. The method further includes establishing the identified safety distance with the at least one control device based upon the identified bunched traffic site. A control device, a computer program and a machine-readable storage medium are further disclosed.

Abstract: Vehicle having dangerous situation notification function and control method, where the method includes monitoring a vicinity of a host vehicle, detecting a movable object in the vicinity of the host vehicle, predicting a dangerous situation in which a collision between the host vehicle and the object is expected, and displaying a warning mark on at least one of a roadway or a wall surface in a projection mapping method in response to the dangerous situation. One or more of an autonomous vehicle, a user terminal, and a server may be in conjunction with an Artificial Intelligence (AI) module, an Unmanned Aerial Vehicle (UAV), an Augmented Reality (AR) device, a Virtual Reality (VR) device, and a device related to a 5G service, etc.

Abstract: A system for a vehicle is provided. The system may include a memory and at least one processor configured to: access a plurality of images of a forward-facing view from the vehicle, the plurality of images corresponding to image data obtained by a camera; determine from the images a first lane marking on a first side of a lane, the lane through which the vehicle can navigate, and a second lane marking on a second side of the lane opposite of the first side; navigate the vehicle autonomously relatively centered between the first and second lane markings; determine from the plurality of images that an object is on the first side or the second side of the lane, and the object beyond the first or second lane marking; and navigate the vehicle autonomously to travel over a driving path that is offset from a center of the lane.

Abstract: The technology relates to using on-board sensor data, off-board information and a deep learning model to classify road wetness and/or to perform a regression analysis on road wetness based on a set of input information. Such information includes on-board and/or off-board signals obtained from one or more sources including on-board perception sensors, other on-board modules, external weather measurement, external weather services, etc. The ground truth includes measurements of water film thickness and/or ice coverage on road surfaces. The ground truth, on-board and off-board signals are used to build the model. The constructed model can be deployed in autonomous vehicles for classifying/regressing the road wetness with on-board and/or off-board signals as the input, without referring to the ground truth.

Abstract: A mobile terminal to be carried by a user of a vehicle acquires a captured image of the vehicle, acquires distance information on a distance to the vehicle based on the captured image, determines whether the distance to the vehicle is within a predetermined allowable distance based on the distance information, and transmits an operation signal corresponding to an operation content input by a user to the vehicle when the distance to the vehicle is determined to be within the allowable distance.

Abstract: Techniques are disclosed for determining a location of an object based at least in part on a motion of the object. The techniques include generating a motion profile based at least in part on motion data received from a mobile device that is associated with the object. The techniques further include receiving, from a camera at a location, a plurality of images that identifies a candidate motion of a candidate object through at least a portion of the location. The techniques further include generating a candidate motion profile corresponding to the candidate motion of the candidate object based at least in part on the plurality of images. Based at least in part on a score generated by comparing the motion profile with the candidate motion profile, the techniques may determine that the candidate object is the object.

Abstract: The technology relates to using on-board sensor data, off-board information and a deep learning model to classify road wetness and/or to perform a regression analysis on road wetness based on a set of input information. Such information includes on-board and/or off-board signals obtained from one or more sources including on-board perception sensors, other on-board modules, external weather measurement, external weather services, etc. The ground truth includes measurements of water film thickness and/or ice coverage on road surfaces. The ground truth, on-board and off-board signals are used to build the model. The constructed model can be deployed in autonomous vehicles for classifying/regressing the road wetness with on-board and/or off-board signals as the input, without referring to the ground truth.

Abstract: A system and method of navigating a vehicle, the vehicle comprising a scanning device and a self-contained navigation system (SCNS) operatively connected to a computer, the method comprising: operating the scanning device for repeatedly executing a scanning operation, each operation includes scanning an area surrounding the vehicle, thereby generating respective scanning output data; operating the computer for generating, based on the scanning output data, a relative map representing at least a part of the area, the map having known dimensions and being relative to a position of the vehicle, wherein the map comprises cells, each cell classified to a class from at least two classes, comprising traversable and non-traversable, and characterized by dimensions equal or larger than an accumulated drift value of the SCNS; wherein non-traversable cells correspond to identified obstacles; receiving SCNS data and updating a position of the vehicle relative to the cells based on the SCNS data.

Abstract: A driving assist device for a vehicle sets a a roadway area ahead of the vehicle, detects an avoidance target existing ahead of the vehicle, and executes collision avoidance control that avoids a collision between the vehicle and the avoidance target. The collision avoidance control is more likely to be executed when the avoidance target is within the roadway area than when the avoidance target is outside the roadway area. The driving assist device detects a roadway end object and a first lane marking of a first lane in which the vehicle exists. An imaginary position is a position apart from the detected position of the roadway end object toward the first lane by a constant distance. The driving assist device sets the imaginary position or the detected position of the the first lane marking as a boundary position of the roadway area based on a predetermined condition.

Abstract: A vehicle control device includes a recognizer configured to recognize situations around a vehicle; a determiner configured to determine any of lanes included in a road on which the vehicle travels as a reference lane; and a driving controller configured to control at least one of a speed and steering of the vehicle according to the situations recognized by the recognizer and the reference lane determined by the determiner, wherein the determiner is configured to, when the vehicle moves from a first road to a second road different from the first road, among a plurality of lanes included in the first road, according to a relative position of a first reference lane determined at a time before the vehicle moves to the second road with respect to the plurality of lanes, determine a second reference lane on the second road.

Abstract: Disclosed is an approach to implement improved anomaly detection. Improved anomaly detection is provided using MSET-SPRT via Monte Carlo simulation that can address problems with conventional MSET-SPRT approaches and provide improved system performance and accuracy.

Abstract: Among other things, we describe techniques for detecting objects in the environment surrounding a vehicle. A computer system is configured to receive a set of measurements from a sensor of a vehicle. The set of measurements includes a plurality of data points that represent a plurality of objects in a 3D space surrounding the vehicle. The system divides the 3D space into a plurality of pillars. The system then assigns each data point of the plurality of data points to a pillar in the plurality of pillars. The system generates a pseudo-image based on the plurality of pillars. The pseudo-image includes, for each pillar of the plurality of pillars, a corresponding feature representation of data points assigned to the pillar. The system detects the plurality of objects based on an analysis of the pseudo-image. The system then operates the vehicle based upon the detecting of the objects.

Abstract: A method and a system for mapping a surroundings of at least one vehicle and for locating the at least one vehicle. Measurement data of the vehicle's surroundings are ascertained by at least one radar sensor of the at least one vehicle, the measurement data of the at least one radar sensor being aggregated. The aggregated measurement data are compared with already existing aggregated measurement data. The aggregated measurement data are optimized by reducing measurement errors based on the comparison between the aggregated measurement data and the already existing aggregated measurement data. A map is generated or updated on the basis of the optimized aggregated measurement data, and the at least one vehicle being located on the generated or updated map by comparing the ascertained measurement data with the generated map.

Abstract: Provided is a system and method that can control a merge of an autonomous vehicle when other vehicles are present on the road. In one example, the method may include iteratively estimating a series of values associated with one or more vehicles in an adjacent lane with respect to an ego vehicle, identifying a trend associated with the one or more vehicles from the iteratively estimated series of values, determining merge intentions of the one or more vehicles with respect to the ego vehicle based on the identified trend over time, verifying the merge intentions against a simulated change in the trend, selecting a merge position of the ego vehicle with respect to the one or more vehicles within the lane based on the verified merge intentions, and executing an instruction to cause the ego vehicle to perform a merge operation based on the selected merge position.

Abstract: Systems and methods for generating mapping data for an autonomous vehicle (e.g., robotic devices). The methods include obtaining three-dimensional environmental data of an environment from a distance sensor. The three-dimensional environmental data includes information relating to one or more objects in the environment. The method further includes identifying at least one planar layer of two-dimensional data from the three-dimensional environmental data to be included in mapping data based on one or more characteristics of an autonomous vehicle, generating mapping data comprising the at least one planar layer of two-dimensional data from the three-dimensional environmental data, and transmitting the mapping data to the autonomous vehicle for use during operation within the environment.

Abstract: The invention is based on controlling operations of an earthmoving machine controllable by the operator by using only one to four controllers and at least one displaying means for displaying controls selectable and controllable by the one to four controllers. The selections and controls of the operations for controlling the earthmoving machine may be transmitted to the control unit both by wire and wirelessly. Thus, the one to four controllers are operable both attached and detached. Further, the control system is arranged to determine the location of the one to four controllers with respect to the earthmoving machine.

Abstract: A position information estimating method and apparatus are provided, wherein the position information estimating method includes estimating initial position information of a vehicle based on motion information of the vehicle that is obtained from one or more sensors, calculating search position information of one or more objects included in an image of surroundings of the vehicle based on the initial position information, evaluating a final reliability of each of the one or more objects based on the initial position information and the search position information, obtaining a filter adaptation parameter corresponding to a weight of the search position information based on the final reliability, and estimating final position information of the vehicle based on the initial position information, the search position information, and the filter adaptation parameter.

Abstract: A vehicular collision avoidance system includes a forward-viewing camera, a rearward-viewing camera, a rearward-sensing non-vision sensor and an electronic control unit. The vehicular collision avoidance system detects vehicles present forward and/or rearward of the equipped vehicle. Responsive to at least one selected from the group consisting of (i) data processing of image data captured by the rearward-viewing camera and (ii) data processing of sensor data captured by the rearward-sensing non-vision sensor, the vehicular collision avoidance system detects another vehicle approaching the equipped vehicle from the rear, determines that the other vehicle is traveling in the same traffic lane as the equipped vehicle, determines speed difference between the vehicles, and determines distance from the equipped vehicle to the other vehicle. Based on such determinations, the system determines that impact with the equipped vehicle by the other vehicle is imminent.

Abstract: In a network of autonomous or semi-autonomous vehicles, an alert may be triggered when one of the vehicles switches from autonomous to manual mode. The alert may be communicated to nearby autonomous vehicles so that drivers of those vehicles may become aware of a potentially unpredictable manual driver nearby. Drivers of autonomous vehicles who may have become disengaged (e.g., sleeping, reading, talking, etc.) during autonomous driving may become re-engaged upon noticing the alert. A re-engaged driver may choose to switch his/her own vehicle from autonomous to manual mode in order to appropriately react to an unpredictable nearby manual driver. In additional or alternative embodiments, the alert may be triggered or intensified when indications of impairment of a nearby driver or malfunction of a nearby vehicle are detected.

Abstract: An autonomous vehicle including a sensor system configured to detect an environment in a field of view of the autonomous vehicle and to output data of the detected environment. A computing system of the autonomous vehicle can detect that a portion of a roadway along a route of the autonomous vehicle in the field of view of the autonomous vehicle is obscured. Responsive to the detection of the obscured portion of the roadway, the computing system can transmit a request for a second autonomous vehicle to view the portion of the roadway. The computing system can receive fill data representative of an output of a second sensor system of the second autonomous vehicle detecting the portion of the roadway in a field of view of the second autonomous vehicle.

Abstract: A vehicular control system includes a plurality of sensors disposed at a vehicle and sensing exterior of the vehicle. An electronic control unit (ECU) includes a processor that processes sensor data captured by the sensors. The ECU, responsive at least in part to processing of captured sensor data as the vehicle travels in a traffic lane of a multi-lane road, determines a rearward approaching vehicle rearward of the equipped vehicle that is in an adjacent traffic lane. The ECU determines a leading vehicle ahead of the equipped vehicle and traveling in the same traffic lane as the equipped vehicle. The ECU controls the equipped vehicle to accelerate the vehicle to at least match the speed of the determined rearward approaching vehicle and to maneuver into the adjacent traffic lane to pass the determined leading vehicle ahead of the determined rearward approaching vehicle.

Abstract: Methods and apparatus to provide accident avoidance information to passengers of autonomous vehicles are disclosed. An example apparatus includes a safety analyzer to determine an autonomously controlled vehicle is in a safe situation at a first point in time and in a dangerous situation at a second point in time. The example apparatus also includes a user interface generator to: generate user interface data to define content for a user interface to be displayed via a screen, the user interface to include a graphical representation of the vehicle, the user interface to graphically indicate the vehicle is in the safe situation at the first point in time; and modify the user interface data so that the user interface graphically indicates the vehicle is in the dangerous situation at the second point in time.

Abstract: In an object detection device to be installed to a vehicle and detect an object outside the vehicle, a position calculator sets multiple candidate points representing a candidate position of the object, based on positions of feature points extracted from a first image captured at a first time. The multiple candidate points are set to be denser within a detection range set based on a distance to the object detected by the ultrasonic sensor than outside the detection range. The position calculator estimates positions of the multiple candidate points at a second time which is after the first time, based on the positions of the multiple candidate points and movement information of the vehicle, and calculates the position of the object by comparing the estimated positions of the multiple candidate points at the second time and the positions of the feature points extracted from a second image captured at the second time.

Abstract: A cloud application may suggest speeds to vehicles based on weather and road conditions in order to improve vehicle safety. Status information associated with a vehicle may be received from one or more computing devices associated with the vehicle. The status information may comprise a location and a speed of the vehicle. A weather condition and a road condition associated with an upcoming location of the vehicle may be obtained. The upcoming location may be determined based at least in part on the received location of the vehicle. A suggested speed at the upcoming location may be determined for the vehicle based at least in part on the weather condition and the road condition. It may be determined that the received speed of the vehicle is greater than the suggested speed. The suggested speed may be sent to the one or more computing devices associated with the vehicle.

Abstract: A distance measuring device includes a motor, a control box and a code discs which are relative rotate driven by the motor. A point position tooth is comprised on the code disc. The control box comprises a distance measuring unit, a detection part and a control unit. The detection part comprises a light emitter and a light receiver which are correspondingly arranged. The control box is rotated relative to the code disc, so that the point position tooth passes through a corresponding position between the light emitter and the light receiver; the control unit receives the signal output of the light receiver, judges the information on alignment status of the point position tooth with the corresponding position, and sends a start or stop operation instruction to the distance measuring unit on the basis of the status information. A method for seeking a distance measuring starting point is also provided.

Abstract: A robotic cleaning apparatus for performing a cleaning operation and a method of cleaning a cleaning space therefor are provided. The method includes acquiring contamination data indicating a contamination level of the cleaning space, acquiring contamination map data based on the contamination data, determining at least one cleaning target area in the cleaning space, based on a current time and the contamination map data, and cleaning the determined at least one cleaning target area. The method and apparatus may relate to artificial intelligence (AI) systems for mimicking functions of human brains, e.g., cognition and decision, by using a machine learning algorithm such as deep learning, and applications thereof.

Abstract: A control system for a vehicle includes a control that determines a planned path of travel for the vehicle along a road being traveled by the vehicle and along a traffic lane in which the vehicle is traveling on the road. The control determines a speed profile for the vehicle along the planned path responsive at least in part to (i) detection of an object along the determined planned path of travel and (ii) determination of a speed limit change along the determined planned path of travel. The control controls the vehicle to maneuver the vehicle along the determined planned path of travel in accordance with the determined speed profile.

Abstract: Aspects of the present disclosure may include methods, apparatuses, and computer readable media for receiving one or more images having a plurality of objects, receiving a notification from an occupant of the self-driving vehicle, generating an attention map highlighting the plurality of objects based on at least one of the one or more images and the notification, and providing at least one of a steering control or a velocity control to operate the self-driving vehicle based on the attention map and the notification.

Abstract: In general, an indication is received through a user interface of an intention of a potential rider to use an autonomous vehicle. In response to the receipt of the indication, a hailing request is sent by a signaling mode to at least one autonomous vehicle that can receive the hailing request directly in accordance with the signaling mode.

Abstract: A vehicular positioning system utilizing multiple optical cameras having contiguous fields of view for reading coded markers having pre-determined positions for determining the position of vehicle inside a structure with a high degree of accuracy. The vehicle positioning system provides for the direct installation and use of a positioning apparatus on a vehicle with a limited number of coded markers to determine the vehicle's position to within millimeter level accuracy.

Abstract: A system and method for real world autonomous vehicle trajectory simulation may include: receiving training data from a data collection system; obtaining ground truth data corresponding to the training data; performing a training phase to train a plurality of trajectory prediction models; and performing a simulation or operational phase to generate a vicinal scenario for each simulated vehicle in an iteration of a simulation. Vicinal scenarios may correspond to different locations, traffic patterns, or environmental conditions being simulated. Vehicle intention data corresponding to a data representation of various types of simulated vehicle or driver intentions.

Abstract: A vehicular control system includes a camera sensor, a radar sensor and a control having a processor that processes image data captured by the camera sensor and radar data captured by the radar sensor to detect another vehicle and to determine an estimated time to arrival of the other vehicle at a location on a road being travelled along by the other vehicle that is in the projected path of travel of the equipped vehicle. Responsive at least in part to (i) determination that the equipped vehicle will arrive at the location before the estimated time to arrival of the other vehicle at the location elapses, and (ii) determination that an object is not present in the projected path of travel of the equipped vehicle, the control determines that it is safe for the equipped vehicle to proceed along the projected path of travel.

Abstract: Provided is a robot, including: a chassis; a set of wheels coupled to the chassis; a processor; and a tangible, non-transitory, machine-readable medium storing instructions that when executed by the processor effectuate operations including: capturing, by an image sensor disposed on a robot, images of a workspace; obtaining, by the processor of the robot or via the cloud, the captured images; comparing, by the processor of the robot or via the cloud, at least one object from the captured images to objects in an object dictionary; identifying, by the processor of the robot or via the cloud, a class to which the at least one object belongs using an object classification unit; and instructing, by the processor of the robot, the robot to execute at least one action based on the object class identified.

Abstract: A motor vehicle power steering mechanism with an electric motor for steering assist and/or steering, and a steering controller, which controls the electric motor with a position control mode for autonomous driving and/or automatic steering and a torque control mode for manual steering by a driver. The steering controller includes a steering column reference controller, an arbitration unit, a column torque controller, a steering algorithm, and the steering system. The steering column reference controller calculates for position control based on a reference position and a measured position, and a first reference steering column torque. A steering algorithm calculates for torque control based on a measured column torque and a second reference steering column torque. The arbitration unit weights and adds the first and second reference steering column torques, and the output of the arbitration unit is input to the column torque controller.

Abstract: Methods and systems for controlling a vehicle. The system includes a localization system, a memory storing a digital map, and an electronic processor. The electronic processor is configured to receive, from the localization system, a current location of the vehicle and determine a future driving segment of the vehicle based on the current location of the vehicle. The electronic processor is further configured to determine at least one performance limitation based on the future driving segment of the vehicle and modify a driving behavior of the vehicle based upon the determined at least one performance limitation.

Abstract: A work machine having an autonomous travel function may include: a cutting section that cuts a work target of the work machine; an image-capturing section that captures an image of the work target cut by the cutting section; and a judging section that judges a state of the cutting section based on the image captured by the image-capturing section. The judging section may judge whether maintenance of or a check on the cutting section is necessary or not based on a result of judgment about the state of the cutting section.

Abstract: Systems and methods for causing a vehicle to avoid an adverse action are described. In one aspect, a sensor on first vehicle may determine the existence and location of lane markers or objects, and use that information to cause a second vehicle to avoid exiting its lane or colliding with an object. Data transmitted from a first vehicle to a second vehicle may be used to determine a path for the second vehicle, such that it avoids an adverse action. This data may include information used to determine a pose of the second vehicle, kinematics of the second vehicle, determine dimensions of the second vehicle, and potential adverse actions. This data may be transmitted while the vehicles are platooning.

Abstract: A method for controlling vehicle system of a vehicle is disclosed. The method comprises determining an expected path of the vehicle, determining a vehicle trajectory for the determined expected path, and determining at least one required control parameter value of a driver assistance system based on the determined vehicle trajectory. Further, the method comprises comparing the at least one required control parameter value to a predefined threshold scheme associated with the driver assistance system, and sending a signal to a Human Machine Interface, HMI, of the vehicle based on the comparison. Then, the method comprises receiving a feedback signal originating from a user of the vehicle, and controlling the driver assistance system based on the comparison and the received feedback signal.

Abstract: Techniques and systems are described that enable evasive steering assist (ESA) with a pre-active phase. An ESA system predicts that a collision with an object is imminent and enters a pre-active phase. The pre-active phase causes a required drop in steering force to occur prior to determining that the collision is imminent. At a later time, the ESA system determines that the collision is imminent and enacts an active phase. The active phase causes a steering force effective to avoid the collision. By enacting the pre-active phase prior to the determination of the imminent collision, the ESA system may provide the additional steering force needed to avoid the collision without delay while simultaneously shielding a driver of vehicle from feeling the drop in steering force.

Abstract: The disclosure includes embodiments for a sensor system for multiple perspective sensor data sets. In some embodiments, a method executed by an ego vehicle includes receiving, from a set of remote vehicles, a set of sensor data describing a set of preliminary heatmaps for a roadway environment. The method includes reconciling discrepancies in the set of heatmaps and a preliminary heatmap of the ego vehicle to form a combined heatmap that describes the objects in the roadway environment as collectively observed by the onboard sensors of the set of remote vehicles and the ego vehicle. The method includes providing the combined heatmap to one or more of the remote vehicles included in the set of remote vehicles. The method includes modifying an operation of the ego vehicle based on the combined heatmap. For example, an operation of an autonomous driving system of the ego vehicle is modified.

Abstract: An apparatus includes a video capture device and a processor. The video capture device may be mounted at a low position. The video capture device may be configured to generate a plurality of video frames that capture a field of view upward from the low position. The processor may be configured to perform video operations to generate dewarped frames from the video frames and detect objects in the dewarped frames, extract data about the objects based on characteristics of the objects determined using said video operations, perform path planning in response the extracted data and generate a video stream based on the dewarped frames. The video operations may generate the dewarped frames to correct a distortion effect caused by capturing the video frames from the low position. The path planning may be used to move the apparatus to a new location. The apparatus may be capable of autonomous movement.

Abstract: The described techniques relate to predicting object behavior based on top-down representations of an environment comprising top-down representations of image features in the environment. For example, a top-down representation may comprise a multi-channel image that includes semantic map information along with additional information for a target object and/or other objects in an environment. A top-down image feature representation may also be a multi-channel image that incorporates various tensors for different image features with channels of the multi-channel image, and may be generated directly from an input image. A prediction component can generate predictions of object behavior based at least in part on the top-down image feature representation, and in some cases, can generate predictions based on the top-down image feature representation together with the additional top-down representation.

Abstract: A point cloud display method includes determining, from a first point cloud, points describing a target object, where the first point cloud describes a surrounding area of a vehicle in which the in-vehicle system is located, and the target object is to be identified by the in-vehicle system, generating a second point cloud based on the points, and displaying the second point cloud.

Abstract: Some embodiments provide an autonomous navigation system which enables autonomous navigation of a vehicle along one or more portions of a driving route based on monitoring, at the vehicle, various features of the route as the vehicle is manually navigated along the route to develop a characterization of the route. The characterization is progressively updated with repeated manual navigations along the route, and autonomous navigation of the route is enabled when a confidence indicator of the characterization meets a threshold indication. Characterizations can be updated in response to the vehicle encountering changes in the route and can include a set of driving rules associated with the route, where the driving rules are developed based on monitoring the navigation of one or more vehicles of the route. Characterizations can be uploaded to a remote system which processes data to develop and refine route characterizations and provide characterizations to one or more vehicles.

Abstract: Systems and methods for self-driving collision prevention are presented. The system comprises a self-driving vehicle safety system, having one or more sensors in communication with a control system. The control system is configured determine safety fields and instruct the sensors to scan a region corresponding to the safety fields. The control system determines exclusion regions, and omits the exclusion regions from the safety field. The safety system may also include capability reduction parameters that can be used to constrain the drive system of the vehicle, for example, by restricting turning radius and speed in accordance with the safety fields.