Abstract: A moving object behavior prediction device improves prediction accuracy and includes: an input map generation unit that generates a single input map in which a region capable of containing a plurality of the moving objects is divided into a plurality of cells, in which each of the cells stores related information of a static object and related information of the moving object; a movement amount estimation unit that estimates a movement amount as a feature amount of each cell from the input map, using a trained convolutional neural network; a movement amount acquisition unit that acquires a movement amount at a current position of the moving object from a movement amount stored in a peripheral cell the moving object; and a future position prediction unit that predicts a future position of the moving object based on a feature amount at a current position of the moving object.

Abstract: A driving assistance device includes a recording unit that records vehicle past traveling positions and related position estimation information, and a relative position estimation unit that estimates a vehicle relative position based on a sensor, adds a history of the relative position to the past traveling position, and uses the output of the sensor. The driving assistance device further includes a position estimation unit that estimates a vehicle position with respect to the past traveling positions and adds a history of the estimated position to the past traveling position, in addition to a deviation determination unit that determines deviation from the past traveling positions using a distribution of the past traveling positions, the vehicle position relative to the past traveling positions, and a reference value. The driving assistance device includes a control unit that controls the vehicle to cancel a warning or deviation when the deviation is determined.

Abstract: Provided is a vehicle control device that can ensure riding comfort or safety. A vehicle control device includes a driving plan calculation unit that calculates a drivable area, which is a space in which an own vehicle is safely drivable, based on a driving environment around the own vehicle and a destination of the own vehicle; and a vehicle motion control unit that calculates a target track including a route and a velocity satisfying a predetermined riding comfort condition in the drivable area calculated by the driving plan calculation unit and controls the own vehicle so as to follow the target track.

Abstract: Provided are a self-map generation device and a self-position estimation device with which it is possible to achieve high-accuracy self-position estimation even when an observation error due to a sensor is large by estimating an error amount of a landmark serving as an observation point and improving the quality of a self-generated point group map.

Abstract: An electronic control device mounted on a vehicle includes: an information acquisition unit that acquires information regarding an environmental element around the vehicle, the environmental element including at least a road surface obstacle that is passable by the vehicle on a road surface; a risk map generation unit that generates a risk map representing a degree of traveling risk of the vehicle at each position around the vehicle based on the information; and a traveling control planning unit that determines a traveling track for traveling control for the vehicle based on the risk map, in which the traveling control planning unit determines the traveling track based on the degree of traveling risk due to the road surface obstacle on the risk map through which a wheel track of the vehicle in the traveling track passes.

Abstract: An object is to provide a map creation/self-location estimation device capable of estimating a location/attitude of an ego vehicle with high accuracy.

Abstract: Realized is a vehicle control device capable of performing automatic driving control that does not reduce turning accuracy of a host vehicle even when an object in a blind spot is different from an assumption and a vehicle speed decreases due to a rapid deceleration. A vehicle control device 10 includes a blind spot object estimation unit 24 that detects a blind spot region 320 of an external-field recognition sensor 12 that recognizes an external field and estimates a blind spot object 310 potential in the blind spot region 320, and a future trajectory generation unit 26 that generates a future trajectory of a host vehicle in consideration of a potential risk from the blind spot object 310 estimated by the blind spot object estimation unit 24 and surrounding information of a vehicle 1 that is the host vehicle.

Abstract: An electronic control device mounted on a vehicle includes: a sensor detectable zone-determining unit that determines a sensor detectable zone representing a zone where an environmental element present around the vehicle is detectable by a sensor mounted on the vehicle based on detection information of the sensor; a cruise control information-generating unit that generates cruise control information for the vehicle based on the detection information of the sensor and the sensor detectable zone determined by the sensor detectable zone-determining unit; and an information output unit that outputs the cruise control information generated by the cruise control information-generating unit.

Abstract: An object of the present invention is to realize lane deviation warning/control at an intersection, on a multilane road and in an environment where no lane dividing lines are present and GNSS accuracy is low.

Abstract: A vehicle control device that autonomously controls a vehicle so as not to cause rapid deceleration that leads to a deterioration in ride quality. The vehicle control device controls first and second deceleration, means that reduce a speed at a deceleration rate large than a deceleration rate of the first deceleration means. The vehicle control device includes a blind spot area detecting unit that detects a blind spot area of a sensor that recognizes an external environment, and a blind spot object estimating unit that estimates a blind spot object that is a virtual moving body hidden in the blind spot area. When a vehicle approaches the blind spot area at a speed reduced by the first deceleration means, the vehicle is decelerated by the second deceleration means when a type of a moving body detected by the sensor is different from a type of the blind spot object.

Abstract: To provide a vehicle control system that is capable of planning a trajectory that can ensure more visibility and enables safe traveling when an invisible range of a sensor exists. A vehicle control system that plans a target trajectory of a vehicle based on recognition information from an external environment sensor, the vehicle control system including a recognizing unit that recognizes an object at a periphery of the vehicle based on the recognition information; and a trajectory planning unit that plans the target trajectory such that an actual detection range of the external environment sensor becomes wide when the recognizing unit recognizes the object.

Abstract: An information processing apparatus having an input device for receiving data, an operation unit for constituting a convolutional neural network for processing data, a storage area for storing data to be used by the operation unit and an output device for outputting a result of the processing. The convolutional neural network is provided with a first intermediate layer for performing a first processing including a first inner product operation and a second intermediate layer for performing a second processing including a second inner product operation, and is configured so that the bit width of first filter data for the first inner product operation and the bit width of second filter data for the second inner product operation are different from each other.

Abstract: An electronic control device mounted on a vehicle includes a blind spot region specifying unit, an information obtaining unit, and a blind spot region dangerous event determining unit. The blind spot region specifying unit specifies a blind spot region that is not included in a detection range of a sensor mounted on the vehicle. The information obtaining unit obtains lane information of a road around the vehicle including the blind spot region. The blind spot region dangerous event determining unit judges assumed behavior of a latent obstacle that possibly exist in the blind spot region based on the lane information of the blind spot region and a positional relationship of the blind spot region on the road with respect to the vehicle.

Abstract: Provided is a vehicle control device that can ensure riding comfort or safety. A vehicle control device includes a driving plan calculation unit that calculates a drivable area, which is a space in which an own vehicle is safely drivable, based on a driving environment around the own vehicle and a destination of the own vehicle; and a vehicle motion control unit that calculates a target track including a route and a velocity satisfying a predetermined riding comfort condition in the drivable area calculated by the driving plan calculation unit and controls the own vehicle so as to follow the target track.

Abstract: A vehicle control system capable of ensuring safety at a low cost even when a control device fails, includes a first control device that implements at least two automatic driving-related functions based on information from external sensors and/or information from a map database, a second control device that implements fewer automatic driving-related functions than the first control device based on the information from the sensors and/or the map database, and a vehicle motion control device that automatically controls a driving state of a host vehicle based on a function planned by the first or second control device including: a backup determination unit that determines whether the future function planned by the first or second control device is backed up by the second control device; and an interface that notifies a driver that system responsibility is switched to the driver, when the backup is not available.

Abstract: A moving object behavior prediction device improves prediction accuracy and includes: an input map generation unit that generates a single input map in which a region capable of containing a plurality of the moving objects is divided into a plurality of cells, in which each of the cells stores related information of a static object and related information of the moving object; a movement amount estimation unit that estimates a movement amount as a feature amount of each cell from the input map, using a trained convolutional neural network; a movement amount acquisition unit that acquires a movement amount at a current position of the moving object from a movement amount stored in a peripheral cell the moving object; and a future position prediction unit that predicts a future position of the moving object based on a feature amount at a current position of the moving object.

Abstract: A correct commutation is realized even if a communication error occurs due to a change in an actuator drive current. An electronic control unit that includes a communication section outputting a control signal and that can transmit the control signal to an actuator connected to the electronic control unit via a power line, includes an actuator operation detection section. When the actuator operation detection section detects an actuator operation, the communication section retransmits the control signal at timing of detecting the actuator operation.

Abstract: A correct commutation is realized even if a communication error occurs due to a change in an actuator drive current. An electronic control unit that includes a communication section outputting a control signal and that can transmit the control signal to an actuator connected to the electronic control unit via a power line, includes an actuator operation detection section. When the actuator operation detection section detects an actuator operation, the communication section retransmits the control signal at timing of detecting the actuator operation.

Abstract: In a processing method using a convolutional neural network, the neural network includes a convolution calculation unit that performs a convolution calculation by using a matrix vector product and a pooling calculation unit that performs a maximum value sampling calculation. A threshold value is set related to the matrix data for the convolution calculation, the matrix data is divided into a first and second halves based on the threshold value. The convolution calculation unit divides a first half convolution calculation by using the first half of the matrix data and a second half convolution calculation by using the second half of the matrix data into two and executes the calculations. The pooling calculation unit selects vector data to which the matrix vector product convolution calculation is to be performed in the second half convolution calculation, along with the maximum value sampling calculation.

Abstract: An information processing apparatus having an input device for receiving data, an operation unit for constituting a convolutional neural network for processing data, a storage area for storing data to be used by the operation unit and an output device for outputting a result of the processing. The convolutional neural network is provided with a first intermediate layer for performing a first processing including a first inner product operation and a second intermediate layer for performing a second processing including a second inner product operation, and is configured so that the bit width of first filter data for the first inner product operation and the bit width of second filter data for the second inner product operation are different from each other.