An acoustic-based Direction of Arrival (DoA) system uses acoustic information to determine the direction of incoming sound, such as a person talking. The direction of the sound is then used to focus a laser-based time of flight (ToF) system to narrow the area of laser illumination, improving the signal to noise ratio because laser illumination is focused on the direction of the sound. The DoA system also provides elevation information pertaining to the source of the sound, to further narrow the required field of view of the laser ToF system.

A plurality of images can be acquired from a plurality of sensors and a plurality of flattened patches can be extracted from the plurality of images. An image location in the plurality of images and a sensor type token identifying a type of sensor used to acquire an image in the plurality of images from which the respective flattened patch was acquired can be added to each of the plurality of flattened patches. The flattened patches can be concatenated into a flat tensor and add a task token indicating a processing task to the flat tensor, wherein the flat tensor is a one-dimensional array that includes two or more types of data. The flat tensor can be input to a first deep neural network that includes a plurality of encoder layers and a plurality of decoder layers and outputs transformer output. The transformer output can be input to a second deep neural network that determines an object prediction indicated by the token and the object predictions can be output.

A plurality of images can be acquired from a plurality of sensors and a plurality of flattened patches can be extracted from the plurality of images. An image location in the plurality of images and a sensor type token identifying a type of sensor used to acquire an image in the plurality of images from which the respective flattened patch was acquired can be added to each of the plurality of flattened patches. The flattened patches can be concatenated into a flat tensor and add a task token indicating a processing task to the flat tensor, wherein the flat tensor is a one-dimensional array that includes two or more types of data. The flat tensor can be input to a first deep neural network that includes a plurality of encoder layers and a plurality of decoder layers and outputs transformer output. The transformer output can be input to a second deep neural network that determines an object prediction indicated by the token and the object predictions can be output.

A method of determining the configuration of a tow load (12) coupleable to a vehicle (10), the method comprising: controlling (116) a vehicle system to obtain an indication of the presence of a unique identifier (40) mounted on the tow load (12); retrieving data (120) encoded within the unique identifier to determine the configuration of the tow load; configuring (124) the vehicle in dependence on the determined configuration of the tow load.

A method for determining a geographical location of a vehicle (10) includes using a camera/sensor device (20) of the vehicle for recording (S10) first image and sensor data (30) from surroundings of the vehicle (10) while the vehicle (10) is traveling a route. The first image and sensor data (30) are assigned geographical coordinates and are sent to a data evaluation unit (50) for creating a digital map. The method continues by using a second camera and sensor device (20) for recording (S40) second image and sensor data (30) from surroundings while the vehicle (10) is traveling the same route and sending (S50) the recorded second image and sensor data (30) to the data evaluation unit (50). The data evaluation unit (50) compares (S60) the recorded second image and sensor data (30) with the digital map of the surroundings (70) and determines (S70) a position of the vehicle (10).

An obstacle shape estimating apparatus and a method thereof, includes: a processor configured to receive a the sensing signal from at least one ultrasonic sensor at a predetermined cycle, to generate positions of one or more obstacles according to distance values of an ultrasonic sensor by estimating the distance values of the ultrasonic sensor based on a sensing signal, and to generate obstacle shape information according to positions of remaining obstacles after deleting a position of an obstacle corresponding to a virtual distance value and a position of an obstacle which does not satisfy a validation condition among the positions of the one or more obstacles; and a storage configured to store data and an algorithm driven by the processor, and the obstacle shape information generated by the processor.

A vehicle control system for an agricultural vehicle including a processing circuit including a processor and memory, the memory having instructions stored thereon that, when executed by the processor, cause the processing circuit to receive (i) image data depicting at least a portion of a hazard area associated with the agricultural vehicle and (ii) ultrasonic sensor data from an ultrasonic sensor monitoring the hazard area, determine, based on a combination of the image data and the ultrasonic sensor data, whether an obstacle is positioned at least partially within the hazard area, and perform an action with respect to the agricultural vehicle in response to a determination that the obstacle is positioned at least partially within the hazard area.

A computer-implemented method comprises: continuously monitoring, by an assisted-driving (AD) system using a sensor, surroundings of a vehicle being controlled by a driver; detecting, by the AD system using the sensor, a parking spot that is available; planning, by the AD system and in response to detecting the parking spot, a trajectory for the vehicle to park in the parking spot; and generating a prompt to the driver, by the AD system and in response to detecting the parking spot, to have the AD system handle parking of the vehicle in the parking spot, the prompt performed before the vehicle reaches the parking spot.

Sensors coupled to a vehicle are calibrated, optionally using a dynamic scene with sensor targets around a motorized turntable that rotates the vehicle to different orientations. One vehicle sensor captures a representation of one feature of a sensor target, while another vehicle sensor captures a representation of a different feature of the sensor target, the two features of the sensor target having known relative positioning on the target. The vehicle generates a transformation that maps the captured representations of the two features to positions around the vehicle based on the known relative positioning of the two features on the target.

A system (1) is provided. The system comprises two primary sensor units (10) and two secondary sensor units (20). The secondary sensor units are configured to receive ultrasonic pulses during time windows, wherein a time window of the time windows comprises a corresponding transmit time of predetermined transmit times. The system is configured to determine a time-of-flight of an ultrasonic pulse of the ultrasonic pulses transmitted at a transmit time of the transmit times based on when the ultrasonic pulse was received during the corresponding time window. The system is further configured to determine a distance between two of the sensor units based on the determined time-of-flight between said sensor units. The system is configured to determine the positions of the sensor units in real-time based on measured movements and the determined distances.