Abstract: In one aspect, a system for disregarding obscured sensor data during the performance of an agricultural operation may include a sensor provided in operative association with an agricultural machine. The sensor may, in turn, be configured to capture three-dimensional data associated with a portion of the field within a field of view of the sensor. A controller of the system may be configured to configured to generate an initial three-dimensional representation of the field based on data received from the sensor. Moreover, the controller may be configured to identify an obscured region within the generated initial three-dimensional representation of the field. Additionally, the controller may be configured to disregard a three-dimensional volume associated with the obscured region from the initial three-dimensional representation of the field to form a modified three-dimensional representation of the field.

Abstract: An artificial intelligence (AI) moving agent is provided. The AI moving agent includes a traveling actuator configured to drive the artificial intelligence moving agent, at least one sensor configured to obtain data for determining an occupied region, and at least one processor configured to determine an occupied region of one or more objects based on the data, and obtain an adjustment zone by resetting a restricted zone set by a user based on the occupied region of the one or more objects.

Abstract: A household appliance control method, device and system, and an intelligent air conditioner are provided. The method includes that: a sound source location is determined by means of a camera; voice information of a user is picked up according to the sound source location; and at least one control operation is performed on a household appliance according to the voice information. The voice information matches at least one corresponding control instruction. By means of the method, a location of at least one user can be detected according to the camera, thereby enhancing at least one pickup audio signal at the location of the at least one user.

Abstract: The present invention relates to a system for steering or guiding a sod harvesting machine (sod harvester) with a high degree of precision, without the need for the sod harvesting machine to have a guide stick or shoe in physical or mechanical contact with the sod to be harvested or to be in connection with remote navigation systems, such as GPS. The system includes a sensor mounted at the front of the harvester and a processor for processing information from the sensor to determine a boundary line and for steering the sod harvester along the boundary line.

Abstract: A method and control system for assessing a potential for collision between a body part of a vehicle and a roadway. The method includes predicting a trajectory path of the vehicle along the roadway, monitoring elevations of the roadway along the trajectory path, and analyzing, at a point along the trajectory path of the roadway, whether the body part of the vehicle will collide with the roadway based upon the estimated elevation of the body part on the vehicle at the point along the trajectory path and the estimated elevation of the roadway at the point along the trajectory path.

Abstract: The present invention provides a control system and a control method capable of appropriately supporting driving of a motorcycle by a rider. The control system includes a control amount setting unit that sets a control amount in auto cruise operation, an execution unit that causes the motorcycle to execute the auto cruise operation, and further including a lane position information acquiring unit that acquires relative position information of a lane boundary with respect to the motorcycle during traveling, and a limitation determining unit that determines to provide a limitation on the auto cruise operation in a case where a determination reference is satisfied, in which the determination reference includes a condition that the lane position information acquired by the lane position information acquiring unit satisfies a prescribed condition.

Abstract: A method for operating a motor vehicle incorporates polling regarding the control of the motor vehicle, leading to an improvement in the working conditions of a plurality of evaluation units accessing sensor units of the motor vehicle. Control commands for controlling the motor vehicle are determined from this polling by a conflict checking unit. The conflict checking unit determines the feasibility of the control commands, taking into consideration predetermined verification criteria with regard to conflicts between the individual control commands and the practicability of the individual control commands. The conflict checking unit also determines a control specification for a vehicle control unit based on the feasibilities and certain decision criteria. Finally, the motor vehicle is controlled by use of the vehicle control unit in accordance with the control specification.

Abstract: Aspects of the disclosure relate to controlling a vehicle having an autonomous driving mode. For instance, that the vehicle is approaching a crosswalk may be determined. A set of segments may be identified for the crosswalk. A set of potential occluded pedestrians may be generated. Each potential occluded pedestrian of the set is assigned a speed characteristic and a segment. The segments of the set of potential occluded pedestrians may be updated over time using the assigned speed characteristics. Sensor data from a perception system of the vehicle is received, and one or more potential occluded pedestrians an having an updated assigned segment corresponding to a segment that is visible to a perception system of the vehicle may be removed from the set of potential occluded pedestrians. After the removing, the set may be used to control the vehicle in the autonomous driving mode.

Abstract: A method for plant treatment, including: receiving a first measurement for a plant from a sensor as the sensor moves within a geographic area comprising a plurality of plants; in response to receipt of the first measurement and prior to receipt of a second measurement for a second plant of the plurality, determining a set of treatment mechanism operation parameters for the plant to optimize a geographic area output parameter based on the first measurement and historical measurements for the geographic area; determining an initial treatment parameter for the plant; and operating a treatment mechanism in a treatment mode based on the set of operating parameters in response to satisfaction of the initial treatment parameter.

Abstract: The present invention provides a control system and a control method capable of appropriately supporting driving of a motorcycle by a rider. The control system includes an execution unit that causes the motorcycle to execute an automatic decelerating operation in a case where an obstacle is located in a predetermined range where a collision avoiding operation of the motorcycle is required, a lane position information acquiring unit that acquires relative position information of a lane boundary with respect to the motorcycle during traveling, and a detection angle range setting unit that set a detection angle range of a forward environment detecting device to be wide in a case where a determination reference is satisfied, in which the determination reference includes a condition that the lane position information acquired by the lane position information acquiring unit satisfies a prescribed condition.

Abstract: An information processing device includes a processor. This processor: obtains at least one of first sensor data output from a first sensor and used to determine an ambient environment of a device in which a third sensor is placed, and second sensor data used to determine an orientation of this device; determines a tilt of a plane in a sensing direction of the third sensor with respect to the orientation of the device based on the at least one of the first sensor data and the second sensor data; determines, in accordance with the tilt determined, a processing target area of third sensor data output from the third sensor and used for object detection processing in the sensing direction of the third sensor; and executes the object detection processing using the processing target area determined, of the third sensor data.

Abstract: The present disclosure discloses an infrared-based road surface monitoring system and method, and an automobile, and relate to the technical field of automated driving of automobile. The infrared-based road surface monitoring system includes: an infrared spotlight assembly, a camera, and an image processor connected with the camera, wherein the infrared spotlight assembly is configured to emit alternate bright and dark infrared stripes, the camera is configured to capture a mode of real-time infrared stripes on a road surface, and the image processor is configured to calculate the form of the stripes on the road surface captured by the camera, so as to obtain road surface or obstacle information and calculate a condition of the road surface, which are fed back to an automatic driving system of an automobile in real time, finally achieving purpose of monitoring road surface and automatic driving.

Abstract: A driving support apparatus for supporting the driving of a vehicle includes: a boundary estimation part for outputting magnetic waves or ultrasonic waves to the periphery of a vehicle, and for estimating the boundary of a roadway based on reflected waves which are detected by a sensor for detecting the reflected waves; a detection part for detecting a mobile body based on the reflected waves which are detected by the sensor; and a region estimation part for estimating a progress region of the object vehicle which is moving in an adjacent lane when this detection part detects the mobile body generated by multiplexed reflections from an object vehicle which is actually moving in the adjacent lane.

Abstract: A smart lawn mower comprises a traveling control module configured to control the traveling and steering of the mower, an image capturing module configured to capture the surrounding images of the mower, an operation module configured to provide a surrounding-determination information, and a storage module configured to store the surrounding-determination information. The operation module determines a grass area by analyzing the surrounding images captured by the image capturing module. The mower defines a grass area accurately without a predetermined boundary.

Abstract: A steering controller includes processing circuitry configured to execute a torque control process that calculates a steering-side operation amount, an angle feedback control process that calculates an angle-side operation amount, and an operation process that operates a drive circuit of an electric motor to adjust torque of the electric motor to a torque command value. The processing circuitry is further configured to execute a switch process that switches, when determining that a controllability of the electric motor is less than or equal to a predetermined efficiency, the torque command value from a value based on the angle-side operation amount to a value based on the steering-side operation amount instead of the angle-side operation amount.

Abstract: A road surface condition prediction system includes a collector which collects pieces of moisture information on moisture on a road surface obtained by detecting the moisture on the road surface of a road on which moving bodies travel, and pieces of position information each indicating a position on the road surface at which the moisture is detected, one or more of the pieces of moisture information and one or more of the pieces of position information being collected from each of the moving bodies; and a predictor which predicts a moisture condition of a target road surface at a time after a time at which moisture on the target road surface is detected, based on moisture information obtained by detecting the moisture on the target road surface, the target road surface being a road surface at a position indicated by at least one of the pieces of position information.

Abstract: A control apparatus for an automatic-driving vehicle includes a load state detecting unit and a collision preventing unit. The load state detecting unit detects a load state of the other vehicle that is present in a periphery of the automatic-driving vehicle. The collision preventing unit performs a collision prevention process based on the load state detected by the load state detecting unit. The collision prevention process is a process for preventing, in advance, collision of an automatic-driving vehicle with a load that may fall onto a road from the other vehicle.

Abstract: The described positional awareness techniques employing sensory data gathering and analysis hardware with reference to specific example implementations implement improvements in the use of sensors, techniques and hardware design that can enable specific embodiments to find new area to cover by a robot encountering an unexpected obstacle traversing an area in which the robot is performing an area coverage task. The sensory data are gathered from an operational camera and one or more auxiliary sensors.

Abstract: The present invention relates to a method for acquiring an object information, the method comprising: obtaining an input image acquired by capturing a sea; obtaining a noise level of the input image; when the noise level indicates a noise lower than a predetermined level, acquiring an object information related to an obstacle included in the input image from the input image by using a first artificial neural network, and when the noise level indicates a noise higher than the predetermined level, obtaining a noise-reduced image of which the environmental noise is reduced from the input image by using a second artificial neural network, and acquiring an object information related to an obstacle included in the sea from the noise-reduced image by using the first artificial neural network.

Abstract: A robot cleaner includes a controller configured to sense an obstacle with a sensing unit of the robot cleaner, and control a driving unit of the robot cleaner to move a main body of the robot cleaner forward into a first contact state in which a front end of a cleaning unit cleans a first region while in contact with the obstacle. After the first contact state, the controller controls the driving unit to move the main body according to a predetermined connecting operation into a second contact state in which the front end of the cleaning unit cleans a second region while in contact with the obstacle, the first region and the second region being arranged along an outline of the obstacle.

Abstract: A materials handling vehicle comprising a path validation tool and a drive unit, steering unit, localization module, and navigation module that cooperate to navigate the vehicle along a warehouse travel path. The tool comprises warehouse layout data, a proposed travel path, vehicle kinematics, and a dynamic vehicle boundary that approximates the vehicle physical periphery. The tool executes path validation logic to determine vehicle pose along the proposed travel path, update the dynamic vehicle boundary to account for changes in vehicle speed and steering angle, determine whether the dynamic vehicle boundary is likely to intersect obstacles represented in the layout data based on the determined vehicle pose at candidate positions along the proposed travel path, determine a degree of potential impingement at the candidate positions by referring to the dynamic vehicle boundary and obstacle data, and modify the proposed travel path to mitigate the degree of potential impingement.

Abstract: A method, apparatus, and system for determining a state of an upcoming traffic light is disclosed. At an autonomous driving vehicle (ADV), an upcoming traffic light ahead in a direction of travel is detected. A relative position of the ADV to the traffic light is determined based on a three-dimensional (3D) position of the traffic light and a position of the ADV. A first image whose content includes the traffic light is captured. A second image of the traffic light is obtained, which comprises cropping the first image and preserving only a first sub-region of the first image that corresponds to the traffic light. One or more third images of the traffic light are retrieved from a precompiled image library based on the relative position of the ADV to the traffic light. A state of the traffic light is determined based on the one or more third images.

Abstract: The disclosure relates to a method for determining a driving instruction, whereby environmental data relevant to the surroundings of the motor vehicle is acquired by at least one detection device of a motor vehicle, after which, as a function of the environmental data, a driving instruction relevant to the driving operation of the motor vehicle is determined, which is specified by a gesture and/or a pose of a traffic controller mapped by the environmental data, wherein the determination of the driving instruction is carried out as a function of at least one set of training data, which was previously acquired by the detection device or a further detection device and which maps a further traffic controller and at least a traffic participant.

Abstract: The disclosure relates to using ambient lighting conditions with passenger and goods pickups and drop offs with autonomous vehicles. For instance, a map of ambient lighting conditions for stopping locations may be generated by receiving ambient lighting condition data for predetermined stopping locations and arranging this data into a plurality of buckets based on time and one of the stopping locations. A vehicle may then be controlled in an autonomous driving mode in order to stop for a passenger by both observing ambient lighting conditions for different stopping locations and, in some instances, also using the map.

Abstract: System, methods, and other embodiments described herein relate to dynamically determining an appropriate responsive action for a moving vehicle in accordance with local traffic regulations. In one embodiment, the disclosed system identifies a stationary vehicle in an environment of the subject vehicle based at least in part on information from a plurality of sensors disposed on the subject vehicle and classifies a type of the stationary vehicle as valid or abandoned based at least in part on the information from the plurality of sensors. A classification of valid indicates that a traffic regulation requires the subject vehicle to undertake a responsive action toward the stationary vehicle. The disclosed system obtains a local traffic regulation based on a location of the subject vehicle and modifies a trajectory of the subject vehicle based on the local traffic regulation when the type of the stationary vehicle is determined to be valid.

Abstract: A method, a device and a computer-readable storage medium with instructions for determining the lateral position of a vehicle relative to the lanes on a roadway. The device has an image processing unit for detecting roadway markings as well as a positioning unit for determining an initial position for the vehicle. An evaluation unit is set up to determine that the detected roadway markings are insufficient for detecting a lateral position of the vehicle by comparing the detected roadway markings with roadway marking information from a lane geometry map for the initial position determined for the vehicle. In this case, the evaluation unit identifies an approximate lateral position of the vehicle while taking into consideration information from the lane geometric map.

Abstract: A vehicle control system includes: an automated driving controller configured to execute automated driving in which at least one of acceleration/deceleration and steering of a vehicle is automatically controlled; and a mode setter configured to set a mode of the automated driving which is executed by the automated driving controller and to set the mode of automated driving to an exchangeable mode in which an occupant who sits in a driver seat is able to be exchanged when the automated driving is executed in a mode in which an occupant needs to sit in the driver seat of the vehicle and predetermined conditions have been satisfied. Accordingly, it is possible to enable smooth exchange of an occupant during automated driving.

Abstract: The present disclosure relates to a method for autonomously controlling a vehicle performed by a vehicle control system, the vehicle control system comprising a zone control system, a collision prediction system and a braking control system, the method comprising the steps of: defining in the zone control system at least a first zone and a second zone relative to a vehicle position, predicting a collision with an obstacle with the prediction system, autonomously braking the vehicle with the braking control system in a first braking mode if the collision is predicted to occur in the first zone and braking the vehicle with the braking control system in a second braking mode if the collision is predicted to occur in the second zone.

Abstract: The vehicle control device includes a start determination unit configured to determine whether or not the inter-vehicle distance between the preceding vehicle and the host vehicle has become equal to or greater than a start threshold greater in value than the stop threshold, and a vehicle control unit configured to start the host vehicle based on a preset speed profile after the start determination unit determines that the inter-vehicle distance between the preceding vehicle and the host vehicle is equal to or greater than the start threshold and maintain the stopped state of the host vehicle until the start determination unit determines that the inter-vehicle distance between the preceding vehicle and the host vehicle is equal to or greater than the start threshold after the preceding vehicle and the host vehicle stop.

Abstract: A driving support apparatus is provided with: a recognizer configured to obtain information indicating a surrounding environment of a host vehicle, and configured to recognize a deceleration required target, which is a target that requires deceleration of the host vehicle; and a supporter configured to perform a driving support operation for supporting the deceleration of the host vehicle, on condition that the deceleration required target is recognized by the recognizer. If a recognition situation, which is recognized by said recognizer, of the deceleration required target that causes the driving support operation to be performed becomes uncertain while performing the driving support operation, said supporter is configured to maintain a preceding operating state of the driving support operation, which is an operating state immediately before the recognition situation becomes uncertain.

Abstract: An autonomous driving system for vehicles includes: an autonomous driving controller for controlling autonomous driving of a host vehicle based on information of nearby vehicles and requesting warning and handover upon determining that a failure in autonomous driving of nearby vehicles has occurred; a communication controller for requesting the warning and handover; selecting an emergency target vehicle and requesting a driving mode of following the emergency target vehicle upon receiving autonomous driving failure information; a human-machine interface device for outputting warning and handover information in response to the request of the autonomous driving controller and the communication controller; and a host vehicle driving controller for controlling driving of the host vehicle in response to the request of the autonomous driving controller and the communication controller.

Abstract: Behavior information input unit (54) receives stop-behavior information about vehicle (100) from automatic-driving control device (30). Image-and-sound output unit (51) outputs inquiry information for inquiring of an occupant whether a possibility of collision between an obstacle and vehicle (100) is to be excluded from a determination object in automatic-driving control device (30) to notification device (2), when a distance from one point on a predictive movement route of the obstacle to the obstacle is greater than or equal to a first threshold, and a speed of the obstacle is less than or equal to a second threshold. Operation signal input unit (50) receives a response signal for excluding the collision possibility from the determination object. Command output unit (55) outputs a command to exclude the collision possibility from the determination object to automatic-driving control device (30).

Abstract: A vehicle control device including: a display; a recognizer; a driving controller that generates a target trajectory of a subject vehicle on the basis of states of objects recognized and controls one or both of a speed and steering of the subject vehicle on the basis of the generated target trajectory; and a display controller that causes to display images resembling the other vehicles recognized as the objects, wherein the display controller causes to display a first image resembling a first vehicle having an influence on a behavior of the subject vehicle according to the driving controller and a second image resembling a second vehicle having an influence on generation of the target trajectory among the other vehicles recognized as the objects with more emphasis than a third image resembling a third vehicle other than the first vehicle and the second vehicle.

Abstract: A display control device for a vehicle include: an acquisition unit that acquires information of an object located at a progress path of the vehicle; and a display control unit that, on the basis of the information acquired by the acquisition unit, causes information of the object to be displayed at a display unit of a spectacles-form wearable terminal, the wearable terminal being provided with the display unit and being configured to be worn by an occupant of the vehicle.

Abstract: An approach is provided for generating a pooled route to extend a service area of a shared vehicle. The approach involves receiving a request from a first user to travel a first route to a destination outside of the service area of the shared vehicle. The approach also comprises determining a second user with a second route that uses the shared vehicle and that is within a threshold proximity, a threshold time, or a combination thereof of the first route, the destination, or a combination thereof. The approach further comprises computing the pooled route for the first user to travel to the destination using the shared vehicle and for the second user to return the shared vehicle to the service area after the first user reaches the destination.

Abstract: A method for carrying out an autonomous driving procedure of a vehicle involves detecting a driver of the vehicle using a camera directed into the interior space of the vehicle. The driver is detected during and after leaving the interior of the vehicle using at least one further camera. A viewing direction of the driver is determined using data detected by the at least one further camera. The autonomous driving procedure is only activated when it is detected that the driver is looking at the vehicle.

Abstract: A vehicle control device (ECU) 10 is configured to, when there is an obstacle 3 in a lane 7, execute traveling course correction processing (S15) of setting a speed distribution zone 40 defining a distribution of an allowable upper limit Vlim of a relative speed of the vehicle 1 with respect to the obstacle 3, and calculate a corrected target traveling course Rc by correcting a target traveling course R so as to prevent the relative speed of the vehicle 1 from exceeding the limit Vlim and enable the vehicle 1 to avoid the obstacle 3. The traveling course correction processing includes restriction processing (S27) of calculating the corrected target traveling course (restricted target travel courses Rc1_r through Rc3_r) such that a lateral avoidance distance (L2_r, L3_r) thereof is restricted to be smaller when a border line of the lane 7 is not detected.

Abstract: The present disclosure provides autonomous vehicle systems and methods that include or otherwise leverage a machine-learned yield model. In particular, the machine-learned yield model can be trained or otherwise configured to receive and process feature data descriptive of objects perceived by the autonomous vehicle and/or the surrounding environment and, in response to receipt of the feature data, provide yield decisions for the autonomous vehicle relative to the objects. For example, a yield decision for a first object can describe a yield behavior for the autonomous vehicle relative to the first object (e.g., yield to the first object or do not yield to the first object). Example objects include traffic signals, additional vehicles, or other objects. The motion of the autonomous vehicle can be controlled in accordance with the yield decisions provided by the machine-learned yield model.

Abstract: Systems and methods for disabling autonomous vehicle input devices are provided. In one example embodiment, a computer implemented method includes identifying an operating mode of an autonomous vehicle. The method includes determining one or more vehicle input devices to be disabled based at least in part on the operating mode of the autonomous vehicle. The vehicle input devices are located onboard the autonomous vehicle. The method includes disabling the one or more vehicle input devices based at least in part on the identified operating mode of the autonomous vehicle such that an input by a user with respect to the one or more vehicle input devices does not affect an operation of the autonomous vehicle.

Abstract: An autonomous driving evaluation apparatus includes: an initial traffic scene setting unit configured to set an initial traffic scene, an initial state of the moving object model, and a road environment in which the autonomous driving vehicle model and the moving object model are disposed; a past traffic scene calculation unit configured to calculate a past traffic scene in which the autonomous driving vehicle model and the moving object model are involved at a past time point back traced from a time point of the initial traffic scene; and a performance evaluation unit configured to evaluate a performance of the autonomous driving algorithm.

Abstract: A vehicle control apparatus and method are provided. The vehicle control apparatus comprises a surrounding detection part, detecting a surrounding status of a host vehicle; and a control part, performing a moving control of the host vehicle according to the surrounding status. In a road system including a first road and a second road intersected with the first road that includes a first lane, a second lane that is an opposite lane of the first lane and a specified area arranged between the first and the second lanes, when the host vehicle enters the first lane, the control part make a determined result whether the host vehicle can enter the specified area or not according to a traffic situation of the second lane, and performs the moving control according to the determined result.

Abstract: A control system for an autonomous agricultural system includes a display configured to display at least one control function associated with at least one operation. The display is configured to output a first signal indicative of a first input. The control system includes a controller comprising a processor and a memory. The controller is communicatively coupled to the display and configured to receive the first signal indicative of the first input and to send a second signal to the display indicative of instructions to display a second control in an unlocked state. The display is configured to output a third signal to the controller indicative of the second input. The controller is configured to receive the third signal and to output a fourth signal indicative of instructions to control the at least one operation of the autonomous agricultural system.

Abstract: A travel-lane determination unit determines whether a vehicle is going to enter a traffic lane on which a toll booth is provided. After it is determined that the vehicle is going to enter the traffic lane, a travel control unit stops the vehicle at a position before the toll booth in a travel direction of the vehicle.

Abstract: A method of autonomously maneuvering a tow vehicle towards a trailer positioned behind the tow vehicle is provided. The method includes receiving one or more images from a camera positioned on a back portion of the tow vehicle. The method also includes receiving, at the data processing hardware, a first user selection of a trailer representation within the one or more images. The method includes sending first instructions to a drive system causing the tow vehicle to autonomously maneuver towards the trailer associated with the first user selection of the trailer representation. The method also includes receiving a second user selection of a coupler representation within the one or more images. The coupler representation associated with a trailer coupler of the trailer. The method also includes sending second instructions to the drive system causing the tow vehicle to autonomously maneuver towards the trailer coupler associated with the second user selection.

Abstract: According to the present invention, there is provided an analysis apparatus (10) including an image analysis unit (11) that detects a predetermined part of a vehicle and persons on the vehicle from each of a plurality of images obtained by imaging the same vehicle from different directions a plurality of times, and detects coordinates of each of the plurality of persons in a coordinate system having the detected predetermined part as a reference; a grouping unit (12) that groups the persons detected from the different images, on the basis of the coordinates; and a counting unit (13) that counts the number of groups.

Abstract: In response to receiving a request from a user device for an on-demand crossing of a roadway at a geo-location, a method determines whether connected vehicles on the roadway are approaching the geo-location. A wait-to-cross-the-roadway message is sent to the user's device. In response to determining connected vehicles are approaching the geo-location, a direction, distance, and speed of connected vehicles relative to the geo-location are determined. Instructions are sent to the connected vehicles to reduce speed and allow the user to cross the roadway, including a message to the user device explaining a purpose of the instructions. In response to determining the one or more connected vehicles comply with the instructions, the one or more processors send a safe-to-cross message to the user, and upon detecting successful roadway crossing of the user device, instructing the connected vehicles to resume travel.

Abstract: A photoelectric conversion device includes a plurality of pixels and a readout circuit. Each of the plurality of pixels includes a pixel circuit including a photoelectric conversion unit. The readout circuit is configured for reading out a signal from the plurality of pixels. The pixel circuit of each of the plurality of pixels includes at least a first transistor that receives a signal based on signal charges generated by the photoelectric conversion unit and is a part of a differential pair. The pixel circuit or the readout circuit has a second transistor. The size of the first transistor and the size of the second transistor are different from each other.

Abstract: Techniques for determining and/or predicting a trajectory of an object by using the appearance of the object, as captured in an image, are discussed herein. Image data, sensor data, and/or a predicted trajectory of the object (e.g., a pedestrian, animal, and the like) may be used to train a machine learning model that can subsequently be provided to, and used by, an autonomous vehicle for operation and navigation. In some implementations, predicted trajectories may be compared to actual trajectories and such comparisons are used as training data for machine learning.

Abstract: Efficient and scalable three-dimensional point cloud segmentation. In an embodiment, a three-dimensional point cloud is segmented by adding points to a spatial hash. For each unseen point, a cluster is generated, the unseen point is added to the cluster and marked as seen, and, for each point that is added to the cluster, the point is set as a reference, a reference threshold metric is computed, all unseen neighbors are identified based on the reference threshold metric, and, for each identified unseen neighbor, the unseen neighbor is marked as seen, a neighbor threshold metric is computed, and the neighbor is added or not added to the cluster based on the neighbor threshold metric. When the cluster reaches a threshold size, it may be added to a cluster list. Objects may be identified based on the cluster list and used to control autonomous system(s).

Abstract: A method for measuring and registering 3D coordinates has a 3D scanner measure a first collection of 3D coordinates of points from a first registration position. A 2D scanner collects horizontal 2D scan sets as 3D measuring device moves from first to second registration positions. A processor determines first and second translation values and a first rotation value based on collected 2D scan sets. 3D scanner measures a second collection of 3D coordinates of points from second registration position. Processor adjusts second collection of points relative to first collection of points based at least in part on first and second translation values and first rotation value. Processor identifies a correspondence among registration targets in first and second collection of 3D coordinates, and uses this correspondence to further adjust the relative position and orientation of first and second collection of 3D coordinates.