Abstract: Provided is a 3D data generating device and the like capable of generating 3D data having a small data size. When generating three-dimensional (3D) data in which a predetermined space is expressed by a plurality of cuboids obtained by partitioning the predetermined space, a position in the height direction for partitioning the predetermined space is decided on the basis of heights corresponding to predetermined objects, and the predetermined space is partitioned at the decided position in the height direction.

Abstract: Disclosed are ultrasound devices and methods for use in guiding a subdermal probe during a medical procedure. A device can be utilized to guide a probe through the probe guide to a subdermal site. In addition, a device can include a detector in communication with a processor. The detector can recognize the location of a target associated with the probe. The processor can utilize the data from the detector and create an image of a virtual probe that can accurately portray the location of the actual probe on a sonogram of a subdermal area. In addition, disclosed systems can include a set of correlation factors in the processor instructions. As such, the virtual probe image can be correlated with the location of the actual probe.

Abstract: Provided herein is a system and method for determining whether a sensor is calibrated and recalibrating of an uncalibrated sensor. The system comprises a sensor system comprising a sensor and an analysis engine configured to determine whether the sensor is uncalibrated. The system further comprises an error handling system configured to determine whether to perform a recalibration in response to the sensor system determining that the sensor is uncalibrated. The error handling system further comprises a recalibration engine configured to perform a recalibration.

Abstract: According to some embodiments, a system receives, at a first sensor of the ADV, a first and a second V2X communication data from a first infrastructure and a second infrastructure respectively. The system determines a first distance from the ADV to the first infrastructure and a second distance from the ADV to the second infrastructure based on the first and the second V2X communication data. The system determines a relative location of the ADV to the first or the second infrastructure based on the first and the second distances and a predetermined distance between the first infrastructure and the second infrastructure. The system retrieves lane information based on the relative location of the ADV to the first or the second infrastructure. The system generates a trajectory based on the lane information to control the ADV autonomously according to the trajectory.

Abstract: A method for planning a trajectory for a self-driving vehicle on a road includes: generating multiple target planned trajectory sets based on information concerning the self-driving vehicle; calculating multiple projected moving ranges and multiple projected moving speeds of an obstacle based on information concerning the obstacle, the projected moving ranges corresponding respectively with multiple time points in a driving time period, the projected moving speeds corresponding respectively with the time points; and selecting one of the target planned trajectory sets as an optimal planned trajectory set based on the target planned trajectory sets, the projected moving ranges and the projected moving speeds for the obstacle.

Abstract: A system includes a machine learning engine configured to receive training data including a plurality of input conditions associated with a state space and a plurality of response maneuvers associated with the state space and train a learning system using the training data and a reward function including a plurality of terms associated with a plurality of end state spaces, each term in the plurality of terms defines an end reward value for each end state space. A value function and policy are generated. The value function comprising a plurality of values, wherein each response maneuvers in the plurality of response maneuvers is associated with a value in the plurality of values related to transitioning from the state space to each end state space, the policy indicative of connections between the state spaces, plurality of values, and the respective end reward value for the plurality of end state spaces.

Abstract: According to some embodiments, an ADV includes a method to plan a path trajectory. The method initializes a path length to a initial path length and performs a path planning operation using the initial path length. The method determines if the path planning operation successfully generates the path trajectory. If the path trajectory fails to be successfully generated, the method determines a reason for the failure. If the path trajectory fails to be generated for reasons other than an obstacle blocking the vehicle, the method defaults the path trajectory to a backup path. If the path trajectory fails to be generated due to an obstacle blocking the vehicle, the method reduces the path length by an incremental amount to generate a reduced path length. The method performs the path planning operation using the reduced path length until a path trajectory is successfully generated or until a stop condition is reached.

Abstract: Dead reckoning combined with GNSS-aided map-matching improves accuracy and reliability of vehicle navigation. A map-match navigation module in a handheld device sends map-match feedback messages to a vehicle state estimator via a port. The module also accepts vehicle speed and inertial navigation data from sensors mounted in the vehicle.

Abstract: A method for object detection and parking spot localization includes receiving data corresponding to detected objects located within a street segment, the data being ascertained by an ascertaining vehicle driving through the street section, determining a detected object distribution, shifting the detected object distribution, and generating a parking map of available parking spots on the street segment based on a parking region distribution and the shifted detected object distribution. The method includes receiving the data, determining the detected object distribution, and shifting of the detected object distribution, for example each time at least one of the at least one ascertaining vehicle drives through the street section.

Abstract: A method for determining an occupancy state of a parking space of a parking facility, including the following method steps: a. detecting magnetic field measured values in the surroundings of the parking space along an axis with the aid of a magnetic field sensor within a predefined duration; b. forming a Poincaré diagram as a function of the magnetic field measured values detected in method step a; c. determining a line of best fit through the points of the Poincaré diagram formed in method step b, with the aid of linear regression; d. determining differences in the vertical direction between the line of best fit and the respective points; e. calculating a mean value of the differences determined in method step d; and f. determining the occupancy state of the parking space as a function of the mean value calculated in method step e.

Abstract: A method, system, and computer program product is provided, for example, for generating heading information. The method comprises receiving a coordinate information for a gantry. The method further comprises receiving an identification information for a link associated with the gantry. Additionally, the method comprises extracting information related to one or more shape points associated with the link based on the identification information. Further, the method comprises matching the coordinate information for the gantry with at least two consecutive shape points from the one or more shape points, wherein the at least two consecutive shape points are closest from the gantry. Also, the method comprises calculating, by a processor, a heading value for the gantry based on the matching. The heading value for the gantry may be used to identify direction information for the gantry, which may further be used for updates.

Abstract: Disclosed is a method for providing input to a vehicle control system to navigate, position and direct a vehicle, by use of a gesture sensing device. The gesture sensing device may be in the form of: a remote, a mobile computing device such as a smartphone, a tablet or smart-watch, or a direction/motion sensor. The vehicle control system with gesture sensing device is useful for the control of radio controlled vehicles, drones and driver and driverless vehicles. In preferred embodiments, the vehicle control system with gesture sensing device is used to control GNSS/GPS enabled electric steer trolling motors, such as motors utilizing the ProNav® Angler, Rhodan® HD GPS® Anchor, Minn-Kota® IPilot, and the MotorGuide® Pinpoint.

Abstract: A vision system of a vehicle includes a camera disposed at a vehicle and having a field of view exterior of the vehicle. The camera may have a wide angle lens with more distortion at side regions of the field of view than at a center region of the field of view. An image processor processes captured image data to determine gradient information of captured image data, and responsive to processing gradient information, determines a potential cross traffic object at one of the side regions of the field of view. The image processor processes multiple frames of captured image data to determine movement of an upper portion and a lower portion of a gradient of the determined potential cross traffic object and, responsive to such processing, determines if the detected potential cross traffic object is a cross traffic object that is moving into a path of the vehicle.

Abstract: An electronic device is provided. The electronic device includes an inertial-measurement unit, an environmental-parameter database, and a computation unit. The inertial-measurement unit is configured to detect inertial information of the electronic device to generate sensor data. The computation unit is configured to perform pose estimation according to the sensor data to obtain a first pose. In response to the electronic device being in a non-moving state, the computation unit performs pose calibration on the first pose according to an environmental parameter in the environmental-parameter database corresponding to a current location at which the electronic device is located.

Abstract: According to an embodiment, a system receives a captured image perceiving one or more lane markers surrounding the ADV from an image capturing device of the ADV. The system detects one or more continuous lane lines based on the one or more lane markers in the captured image by applying a machine learning model to the captured image, where the machine learning model includes a number of layers of nodes and the machine learning model includes a weighted softmax cross-entropy loss within at least one of the layers in training. The system generates a trajectory based on the one or more continuous (e.g., whole) lane lines to control the ADV autonomously according to the trajectory.

Abstract: Disclosed is method of computing a round trip delay between a pair of nodes, the method comprising transmitting at least one beacon at a known transmit time from each of the nodes; measuring the times-of-arrival of the beacons at other of the nodes; and estimating a round trip delay between the nodes from the measured times-of-arrival and the transmit times; and correcting the round trip delay for either or both of a frequency offset between the nodes and relative motion between the nodes.

Abstract: A system for providing drug library data to a medical device (31, 32) located within a healthcare environment, the system comprises a local network (33) of the healthcare environment (3) and at least one medical device (31, 32) for administering a drug to a patient, the at least one medical device (31, 32) being located in the healthcare environment (3) and connected to the local network (33). Herein, a drug library server (1) is connected to the local network (3) of the healthcare environment (3) via a public communication network (2) and constituted to provide drug library data to the at least one medical device (31, 32) via the public communication network (2). In this way, a system and method for providing drug library data to a medical device located within a healthcare environment is provided which allows for an easy creation, editing and sharing of drug library data for use and distribution between medical devices within a particular healthcare environment and across different healthcare environments.

Abstract: Provided is a method including detecting at least one external vehicle located within a predetermined distance from a first vehicle using at least one sensor; determining a risk due to the at least one external vehicle based on a type of the at least one external vehicle; and planning a traveling path of the first vehicle based on the risk due to the at least one external vehicle.

Abstract: A method for transferring operational data to a medical device (31, 32) for administering a drug to a patient, the medical device (31, 32) being located in a healthcare environment (3) and connected to a local network (33), comprises the steps of: transferring operational data to be transferred to the medical device (31, 32) to a communication device (40, 41); reading, by means of the communication device (40, 41), information from an identification tag (8) attached to the medical device (31, 32), wherein the information causes the communication device (40, 41) to access a communication interface (310, 320, 340) associated with the medical device (31, 32); and transferring the operational data from the communication device (40, 41) via the communication interface (310, 320, 340) to the medical device (31, 32).

Abstract: A robotic mapping and tracking system including a robot and boundary posts are disclosed. The robot includes an ultrasonic transmitter, a processor and a camera component. The boundary posts are configured to be placed adjacent to a boundary of a working region. Each boundary post of the plurality of boundary posts includes an ultrasonic receiver. Time-of-flights of the ultrasonic waves are measured to identify distances in between the robot and boundary posts. The camera component of the robot captures an image of an environment of the robot. The processor of the robot analyzes the image of the environment and identifies at least a portion of the working region in front of the robot from the image. The processor of the robot determines a moving route based on the identified portion of the working region in front of the robot and the distances in between the robot and the boundary posts.

Abstract: A control method in a mobile automation apparatus includes: storing a plurality of parametric path definition identifiers and respective corresponding sub-region identifiers identifying sub-regions in a facility; receiving a task command containing (i) a task identifier, and (ii) a target one of the sub-region identifiers identifying a target one of the sub-regions; retrieving a selected one of the parametric path definition identifiers corresponding to the target sub-region identifier; obtaining a selected one of a plurality of parametric path definitions—including a shape parameter and a reference parameter name—corresponding to the selected parametric path definition identifier; obtaining a reference parameter value, and generating a sequence of mobile automation apparatus poses according to the parametric path definition; each pose corresponding to a location in the facility; and controlling a locomotive assembly of the mobile automation apparatus to traverse the target sub-region according to the sequenc

Abstract: Certain aspects of the present disclosure provide techniques for generating geofences for an address using an iterative process. An example method generally includes receiving an address to generate a geofence around. A position associated with the received address is obtained from a map service. A set of perimeter points is generated for the received address. Each perimeter point is associated with a latitude and longitude position determined based on a traversal direction and an iteratively determined distance from the obtained position associated with the received address. A geofence is generated for the received address based on the set of perimeter points.

Abstract: An autonomous vehicle is operable to follow a primary trajectory that forms a portion of a route. While controlling the autonomous vehicle, the autonomous vehicle calculates a failsafe trajectory to follow as a response to a predetermined type of event.

Abstract: A mobile application executable on at least one mobile device for associating actions with a plurality of independent geographic locations and a system for implementing the actions are provided. The system and mobile application carry out a method including the steps of configuring a graphical user interface to include a map and requesting the user to select a designated pin location. The system includes a main controller configured to communicate with the at least one mobile device and initiate the actions based on the location of the at least one mobile device relative to the plurality of independent geographic locations. At least one database is in communication with the main controller and accessible to the mobile and away services module for storing data related to the actions associated with the plurality of independent geographic locations and the at least one mobile device.

Abstract: The above measurement device acquires a distance from a vehicle to a partial area of a feature and an angle formed by a direction of viewing the partial area from the vehicle and a traveling direction of the vehicle, for a first area and a second area, the first area being the partial area at a first time, the second area being the partial area different from the first area at a second time. Also, the measurement device acquires length information including a width of the feature. Then, the measurement device calculates a speed of the vehicle from the first time to the second time based on acquisition results by the first acquisition unit and the second acquisition unit. Thus, the speed of the vehicle may be calculated by a scan of a feature in one period.

Abstract: A method for performing a cleaning operation with a cleaning device includes: determining whether the cleaning device and a terminal are located in a same general area; acquiring a first geographical location at which the terminal is currently located; and performing the cleaning operation according to the first geographical location and a specified cleaning mode.

Abstract: Method and apparatus are disclosed for mobile device tethering for a remote parking assist system of a vehicle. An example vehicle includes a plurality of wireless nodes, occupant detection sensors, and a body control module. The body control module determines an initial exit location relative to the vehicle of a mobile device based on signal received from the wireless nodes and the occupant detection sensors. The body control module also tracks a location of the mobile device based on the initial exit location using dead reckoning and enables remote parking assist when the location is within a threshold distance.

Abstract: A parking assist apparatus includes a one-way traffic determiner that determines whether or not a vehicle is located on a one-way road, and a parking processor that, when it is determined that the vehicle is located on the one-way road case, does not perform a vehicle control for parallel parking including a path through which the vehicle reverses.

Abstract: Examples provide a customized travel expense system for predicting travel expenses for a user planning to travel to a remote training location and disbursing an authorized portion of the predicted travel expenses to the user at a predetermined time for utilization during a training event. A mode of travel is selected based on mode selection criteria and training schedule data. The total predicted travel expenses and a daily expense allotment is calculated based on the selected mode and a rate associated with the selected mode. A user's presence at a training event provided at the remote training location is verified based on sensor data obtained from sensor devices located at the training location. An amount of funds equivalent to at least a portion of the total predicted travel expenses is uploaded to an account associated with the user in real-time for utilization by the user during the training event.

Abstract: In one embodiment, a system of an ADV perceives a driving environment surrounding the ADV using a plurality of sensors mounted on the ADV. The system identifies a blind spot based on the perceived driving environment surrounding the ADV. The system in response to identifying the blind spot, receives an image having the blind spot from an image capturing device disposed within a predetermined proximity of the blind spot. In some embodiments, the system receives the image having the blind spot from a remote server communicatively coupled to the image capturing device. The system identifies an obstacle of interest at the blind spot of the ADV based on the image. The system generates a trajectory based on the obstacle of interest at the blind spot to control the ADV to avoid the obstacle of interest.

Abstract: A map-data collection system for mapping an area includes a first sensor, a receiver, and a controller-circuit. The first-sensor is for installation on a first-vehicle. The first-sensor is configured to gather perception-data of an area from a first-perspective. The receiver is for installation on the first-vehicle. The receiver is configured to receive perception-data gathered by a second-sensor mounted on a second-vehicle proximate to the first-vehicle. The second-sensor is configured to gather perception-data of the area from a second-perspective different from the first-perspective. The controller-circuit is in communication with the first-sensor and the receiver. The controller-circuit is configured to determine composite-data in accordance with the perception-data from the first-sensor on the first-vehicle and the perception-data from the second-sensor on the second-vehicle.

Abstract: A system for providing adaptive cruise control in a motorcycle. The system includes an electronic controller configured to determine the presence of a vehicle on one side of a direct path of travel of the motorcycle based on data received from a transceiver, the vehicle within a field of view of the transceiver. The electronic controller locks the motorcycle with the vehicle and dynamically controls the speed of the motorcycle based on an output of a kinematic controller, wherein the kinematic controller configured to receive an input including at least one of an item selected from the group consisting of distance of the motorcycle to the vehicle, velocity of the vehicle, velocity of the motorcycle, a cruise set speed associated with the motorcycle, a desired separation distance between the motorcycle and the vehicle, and a desired separation time between the motorcycle and the vehicle.

Abstract: A parking assisting apparatus includes an imaging unit acquiring image information corresponding to an image of surroundings of a vehicle, an image processing section recognizing a feature shape in the image by processing the image information, an obstacle detecting section acquiring positional-relationship information corresponding to a positional relationship between the vehicle and an obstacle present around a parking space, and a manner-of-parking selecting section selecting a manner of parking the vehicle in the parking space from manner candidates including perpendicular parking and parallel parking based on the feature shape and the positional-relationship information.

Abstract: A device for locating and operating a medical device having different operational modes is provided. The device comprises a first database, wherein from a plurality of different operational modes for the medical device and from a plurality of different geographical positions of use for operation of the medical device, a predetermined operational mode is assigned to a geographical position of use. A positioning means is adapted to be provided at the medical device and to determine the geographical position of the medical device, and a control means is connected to the positioning means and the database and is adapted to determine, by mapping the determined geographical position of the medical device with the geographical positions of use stored in the first database, the predetermined operational mode which is assigned to the geographical position where the medical device is located, and to set the medical device to said predetermined operational mode.

Abstract: The present invention relates to a method of operating a scanner for monitoring an least two-dimensional protected field that is divided into a plurality of cyclically scannable monitored fields, wherein the scanner comprises at least one light transmitter and at least one light receiver. The invention further relates to a scanner for monitoring an at least two-dimensional protected field.

Abstract: A vehicle equipped with a driving assistance system with a low power mode, the vehicle includes an image capturing device and a processor. The image capturing device capture external scenes outside of the vehicle to generate a series of observing images. In a low power mode, the processor identifies a second vehicle and a plurality of features on exterior of the second vehicle, determines a reference vector corresponding to a travel direction of the second vehicle according to at least two features, and determines if a change of a detection value relating to the reference vector in the series of observing images exceeds a predetermined threshold. When the reference vector exceeds the predetermined threshold, the processor determines that the second vehicle has changed the travel direction.

Abstract: A navigation ECU includes a route calculator section, an automated driving distance calculator section, and a display processor section. The route calculator section calculates a plurality of route candidates to a destination designated by a user. The automated driving distance calculator section calculates an automated driving enabled distance that is a distance of a road segment, where a vehicle is able to be run by an automated driving, with respect to each of the plurality of route candidates. The display processor section displays a route selection screen view containing the information which indicates an automated driving enabled distance versus a full length, with respect to each of the plurality of route candidates.

Abstract: A communication system for vehicles includes a first communication device disposed at a trailer and a second communication device disposed at a vehicle. The communication device wirelessly transmits a communication to the second communication device, and the second communication device receives the transmitted communication from the first communication device. Responsive to processing of the transmitted communication received by the second communication device, the communication system determines an angle of the trailer relative to the vehicle. The first and second communication devices may include first and second dedicated short range communication devices. One of the devices may include spaced apart antennae, whereby the system may determine the angle of the trailer via triangulation based on an antenna of one of the communication devices and the spaced apart antennae of the other of the communication devices.

Abstract: A magnetic sensing apparatus for sensing and tracking the position of a target that is independent of the rotation of the target about the magnetic sensing elements. The apparatus provides increased resolution and increased sensor element to sensor element spacing. Also, the apparatus reduces the susceptibility to external interference from interaction with external magnetic fields.

Abstract: Geo-fences may be dynamically reconfigured to identify merchants that may be of particular interest to users. In some instances, a geo-fence may be defined for a particular user based on a variety of information, in order to identify merchants that may be of interest to the user and that are located within proximity to the user. The geo-fence may be defined based on purchase history, places that are frequented by the user, a mode of transportation, a navigation or transportation route, user preferences, and/or a variety of other information. Information regarding the identified merchants may be displayed or otherwise output to notify the user of nearby merchants that may be of interest.

Abstract: The technology relates to determining general weather conditions affecting the roadway around a vehicle, and how such conditions may impact driving and route planning for the vehicle when operating in an autonomous mode. For instance, the on-board sensor system may detect whether the road is generally icy as opposed to a small ice patch on a specific portion of the road surface. The system may also evaluate specific driving actions taken by the vehicle and/or other nearby vehicles. Based on such information, the vehicle's control system is able to use the resultant information to select an appropriate braking level or braking strategy. As a result, the system can detect and respond to different levels of adverse weather conditions. The on-board computer system may share road condition information with nearby vehicles and with remote assistance, so that it may be employed with broader fleet planning operations.

Abstract: A system includes a vehicle-processor in communication with an external vehicle sensor, wherein the processor is configured to detect another vehicle within a predefined proximity. The processor is further configured to instruct the external vehicle sensor to scan a characteristic of the another vehicle, responsive to the detecting. Also, the processor is configured to determine if the characteristic includes an anomaly, responsive to the scan and notify the driver of the another vehicle of the anomaly, responsive to the determination.

Abstract: Disclosed herein is a vehicle control apparatus and a vehicle control method. The vehicle control apparatus includes a first sensor configured to detect an object; a second sensor configured to detect driver steering input information; an Autonomous Emergency Steering Assist (AESA) device configured to perform automatic emergency steering assist; a Driver-initiated Emergency Steering Assist (DESA) device configured to perform emergency steering assistance based on the driver steering input information; and a controller configured to control the AESA device to apply a steering assist control torque value corresponding to the AESA device differently for each of collision risk levels in a collision risk situation with the detected object, and when there is the detected driver steering input information, to control the DESA device to apply a steering assist control torque value corresponding to the DESA device according to the driver steering input information.

Abstract: A device and related method are associated with a focus vehicle. The device comprises a sensor that senses area information about an information area usable to assist the focus vehicle and an oncoming vehicle to pass each other, and a transmitter that broadcasts the area information via a vehicle-to-vehicle network (V2VN) protocol. The device comprises a controller that is configured to collect, store, and share, via the V2VN, area information comprising passing information when on a narrow road, obstacle information when on a wide road, and passing obstacle information when on the narrow road. The device produces an assisting result from the area information and vehicle information related to the focus vehicle and the oncoming vehicle, and a display that displays virtual traffic information representing at least stopping and proceeding of the focus vehicle onto a windshield of the vehicle.

Abstract: A through the road (TTR) hybridization strategy is proposed to facilitate introduction of hybrid electric vehicle technology in a significant portion of current and expected trucking fleets. In some cases, the technologies can be retrofitted onto an existing vehicle (e.g., a truck, a tractor unit, a trailer, a tractor-trailer configuration, at a tandem, etc.). In some cases, the technologies can be built into new vehicles. In some cases, one vehicle may be built or retrofitted to operate in tandem with another and provide the hybridization benefits contemplated herein. By supplementing motive forces delivered through a primary drivetrain and fuel-fed engine with supplemental torque delivered at one or more electrically-powered drive axles, improvements in overall fuel efficiency and performance may be delivered, typically without significant redesign of existing components and systems that have been proven in the trucking industry.

Abstract: A vehicle control apparatus includes a deviation detector and a motor controller. The deviation detector is configured to detect a deviation of a host vehicle from a travel lane. Upon the deviation detector detecting the deviation, the motor controller is configured to increase and decrease torque of a driving motor that transfers a driving force to a wheel.

Abstract: The present disclosure provides an autonomous vehicle site test method and apparatus, a device and a readable medium. The method comprises: in a site test, collecting state information of the autonomous vehicle in real time; according to the state information of the autonomous vehicle and a pre-obtained test demand, obtaining deployment information of a simulated obstacle needed in the test, wherein the deployment information of the simulated obstacle includes a simulated obstacle class, a deployment location of the simulated obstacle and a predetermined action of the simulated obstacle; according to the deployment information of the simulated obstacle, controlling the simulated obstacle corresponding to the simulated obstacle class to perform a predetermined action at the deployment location, to test the autonomous vehicle's response performance to the deployed simulated obstacle.

Abstract: A wireless charging system for a vehicle is described, as well as methods of using the system. The system may include a computer programmed with instructions stored in memory and executable by a processor, and a radar device controlled by the computer. One method includes: aligning a reception coil on a vehicle with a charging coil using ground penetrating radar data; and then receiving inductive charge at the reception coil.

Abstract: According to some embodiments, a system generates a driving trajectory from a starting point to a destination point for an ADV. In one embodiment, the system calculates a first trajectory based on a map and a route information. The system generates a path profile based on the first trajectory, traffic rules, and blocking obstacles perceived by the ADV. The system determines non-blocking obstacles perceived by the ADV. The system generates a speed profile of the ADV for the path profile based on the non-blocking obstacles to identify a speed of the ADV to avoid the blocking obstacles in view of the non-blocking obstacles. The system generates a second trajectory based on the path profile and the speed profile to control the ADV autonomously according to the second trajectory.

Abstract: A direction identification unit (21) identifies a traveling direction in which a surrounding vehicle travels in a partial region of a region indicated by image information (42) obtained by photographing using a camera (31). A feature amount acquisition unit (22) acquires a reference feature amount (41) being a feature amount computed from a reference image corresponding to the identified traveling direction. A vehicle determination unit (23) computes an image feature amount being a feature amount of the image information of the partial region and compares the computed image feature amount with the acquired reference feature amount (41), thereby determining whether the surrounding vehicle is present in the partial region.