A vehicle control device includes a peripheral recognition unit configured to recognize a peripheral status of a vehicle including a position of a traffic participant present in a periphery of the vehicle on the basis of an output of an in-vehicle device, an estimation unit configured to estimate a peripheral attention ability of the traffic participant on the basis of an output of the in-vehicle device, and a risk area setting unit configured to set a risk area of the traffic participant on the basis of a result of the estimation performed by the estimation unit.

A vehicle control device includes a recognizer configured to recognize a surrounding environment of a vehicle including a moving object present around the vehicle and a controller configured to control at least one of a speed and steering of the vehicle. The controller restricts access to the moving object when the moving object is rotating around a vertical axis at a speed greater than or equal to a threshold value so that a front surface of the moving object recognized by the recognizer faces a position interfering with a position in a traveling direction of the vehicle as compared with when the moving object is not rotating.

A vehicle control device includes a recognizer configured to recognize a surrounding environment of a vehicle, a setter configured to set a first risk area in a surrounding area of the vehicle on the basis of a recognition result of the recognizer, and a controller configured to control at least one of a speed and steering of the vehicle. The setter sets the first risk area so that the first risk area includes an area between the moving object and a first end of a crosswalk where the moving object is scheduled to arrive in the crosswalk when the moving object is entering the crosswalk which is provided in front of the vehicle and where the vehicle is scheduled to pass on the basis of the recognition result of the recognizer. The controller prevents the vehicle from entering the first risk area when a first predetermined condition is satisfied.

A device includes a storage device configured to store a program; and a hardware processor, wherein, the hardware processor executes the program stored in the storage device to: recognize positions of a plurality of traffic participants; determine a temporary goal which each of the plurality of traffic participants is trying to reach in the future, based on the recognition results; and simulate a movement process in which each of the plurality of traffic participants moves toward the temporary goal using a movement model to estimate an action in the future of each of the plurality of traffic participants.

A method, an apparatus and a system for displaying an image of a vehicle blind spot are provided. The apparatus is coupled to at least one external camera disposed outside the vehicle, at least one internal camera disposed inside the vehicle with a lens facing towards a driver, and at least one display disposed inside the vehicle. The method comprises capturing an image of the external environment of the vehicle by the external camera as an external image, capturing an image including the driver by the internal camera as an internal image, recognizing a face of the driver and a displacement of the face in the internal image, and adjusting a position of an ROI in the external image according to the recognized displacement, to display an image of the ROI on the display corresponding to the external camera. The displayed image of the external of the vehicle may be adjusted according to the posture or the angle of view of the driver, to have the image of the current blind spot by the driver shown correctly.

Systems and methods for lane marking and road sign recognition are provided. The system aligns image level features between a source domain and a target domain based on an adversarial learning process while training a domain discriminator. The target domain includes one or more road scenes having lane markings and road signs. The system selects, using the domain discriminator, unlabeled samples from the target domain that are far away from existing annotated samples from the target domain. The system selects, based on a prediction score of each of the unlabeled samples, samples with lower prediction scores. The system annotates the samples with the lower prediction scores.

A method for providing vehicle alerts includes receiving an ignition signal indicating a current status of an ignition of a vehicle and receiving a gear position signal indicating a current gear position of a transmission of the vehicle. The method also includes receiving a vehicle speed signal indicating a current vehicle speed of the vehicle and identifying a vehicle alert data file based on at least the ignition signal, the gear position signal, the vehicle speed signal. The method also includes retrieving the vehicle alert data file from a vehicle alert database and loading data associated with the vehicle alert data file into a buffer. The method also includes outputting contents of the buffer to at least one output device of the vehicle.

A system uses neural networks to determine intents of traffic entities (e.g., pedestrians, bicycles, vehicles) in an environment surrounding a vehicle (e.g., an autonomous vehicle) and generates commands to control the vehicle based on the determined intents. The system receives images of the environment captured by sensors on the vehicle, and processes the images using neural network models to determine overall intents or predicted actions of the one or more traffic entities within the images. The system generates commands to control the vehicle based on the determined overall intents of the traffic entities.

Systems and methods for obstacle detection are provided. The system aligns image level features between a source domain and a target domain based on an adversarial learning process while training a domain discriminator. The target domain includes one or more road scenes having obstacles. The system selects, using the domain discriminator, unlabeled samples from the target domain that are far away from existing annotated samples from the target domain. The system selects, based on a prediction score of each of the unlabeled samples, samples with lower prediction scores. The system annotates the samples with the lower prediction scores.

A system for method for future forecasting using action priors that include receiving image data associated with a surrounding environment of an ego vehicle and dynamic data associated with dynamic operation of the ego vehicle. The system and method also include analyzing the image data and detecting actions associated with agents located within the surrounding environment of the ego vehicle and analyzing the dynamic data and processing an ego motion history of the ego vehicle. The system and method further include predicting future trajectories of the agents located within the surrounding environment of the ego vehicle and a future ego motion of the ego vehicle within the surrounding environment of the ego vehicle.