Abstract: Techniques are discussed for segmenting sensor data captured by a sensor to remove a ground surface from the sensor data. A first technique includes capturing sensor data represented as multichannel image and segmenting the image according to image processing techniques. The ground surface can be removed from the sensor data, and a subset of the sensor data can be associated with a voxel space. A second technique includes capturing sensor data and unprojecting the sensor data to generate three dimensional data, which can be associated with a voxel space. Ground plane data associated with a location can be accessed or determined and voxel data that is within a threshold height of the ground plane data can be removed from the voxel space. Clustering techniques can determine objects represented in the data, and a vehicle can be controlled based on the objects.

Abstract: A vehicle control system includes various sensors. The system can include, among others, LIDAR, RADAR, SONAR, cameras, microphones, GPS, and infrared systems for monitoring and detecting environmental conditions. In some implementations, one or more of these sensors may become miscalibrated. Using data collected by the sensors, the system can detect a miscalibrated sensor and generate an indication that one or more sensors have become miscalibrated. For example, data captured by a sensor can be processed to determine an average height represented by the sensor data and compared to an average height of data captured by other sensors. Based on a difference in heights, an indication can be generated identifying a miscalibrated sensor.

Abstract: A system may include one or more processors configured to receive a plurality of images representing an environment. The images may include image data generated by an image capture device. The processors may also be configured to transmit the image data to an image segmentation network configured to segment the images. The processors may also be configured to receive sensor data associated with the environment including sensor data generated by a sensor of a type different than an image capture device. The processors may be configured to associate the sensor data with segmented images to create a training dataset. The processors may be configured to transmit the training dataset to a machine learning network configured to run a sensor data segmentation model, and train the sensor data segmentation model using the training dataset, such that the sensor data segmentation model is configured to segment sensor data.

Abstract: A vehicle can include various sensors to detect objects in an environment. Sensor data can be captured by a perception system in a vehicle and represented in a voxel space. Operations may include analyzing the data from a top-down perspective. From this perspective, techniques can associate and generate masks that represent objects in the voxel space. Through manipulation of the regions of the masks, the sensor data and/or voxels associated with the masks can be clustered or otherwise grouped to segment data associated with the objects.

Abstract: A vehicle control system includes various sensors. The system can include, among others, LIDAR, RADAR, SONAR, cameras, microphones, GPS, and infrared systems for monitoring and detecting environmental conditions. In some implementations, one or more of these sensors may become miscalibrated. Using data collected by the sensors, the system can detect a miscalibrated sensor and generate an indication that one or more sensors have become miscalibrated. For example, data captured by a sensor can be processed to determine an average height represented by the sensor data and compared to an average height of data captured by other sensors. Based on a difference in heights, an indication can be generated identifying a miscalibrated sensor.

Abstract: A vehicle can include various sensors to detect objects in an environment. Sensor data can be captured by a perception system in a vehicle and represented in a voxel space. Operations may include analyzing the data from a top-down perspective. From this perspective, techniques can associate and generate masks that represent objects in the voxel space. Through manipulation of the regions of the masks, the sensor data and/or voxels associated with the masks can be clustered or otherwise grouped to segment data associated with the objects.

Abstract: A system may include one or more processors configured to receive a plurality of images representing an environment. The images may include image data generated by an image capture device. The processors may also be configured to transmit the image data to an image segmentation network configured to segment the images. The processors may also be configured to receive sensor data associated with the environment including sensor data generated by a sensor of a type different than an image capture device. The processors may be configured to associate the sensor data with segmented images to create a training dataset. The processors may be configured to transmit the training dataset to a machine learning network configured to run a sensor data segmentation model, and train the sensor data segmentation model using the training dataset, such that the sensor data segmentation model is configured to segment sensor data.

Abstract: A system may include one or more processors configured to receive a plurality of images representing an environment. The images may include image data generated by an image capture device. The processors may also be configured to transmit the image data to an image segmentation network configured to segment the images. The processors may also be configured to receive sensor data associated with the environment including sensor data generated by a sensor of a type different than an image capture device. The processors may be configured to associate the sensor data with segmented images to create a training dataset. The processors may be configured to transmit the training dataset to a machine learning network configured to run a sensor data segmentation model, and train the sensor data segmentation model using the training dataset, such that the sensor data segmentation model is configured to segment sensor data.

Abstract: A vehicle can include various sensors to detect objects in an environment. Sensor data can be captured by a perception system in a vehicle and represented in a voxel space. Operations may include analyzing the data from a top-down perspective. From this perspective, techniques can associate and generate masks that represent objects in the voxel space. Through manipulation of the regions of the masks, the sensor data and/or voxels associated with the masks can be clustered or otherwise grouped to segment data associated with the objects.

Abstract: A vehicle control system includes various sensors. The system can include, among others, LIDAR, RADAR, SONAR, cameras, microphones, GPS, and infrared systems for monitoring and detecting environmental conditions. In some implementations, one or more of these sensors may become miscalibrated. Using data collected by the sensors, the system can detect a miscalibrated sensor and generate an indication that one or more sensors have become miscalibrated. For example, data captured by a sensor can be processed to determine an average height represented by the sensor data and compared to an average height of data captured by other sensors. Based on a difference in heights, an indication can be generated identifying a miscalibrated sensor.

Abstract: A system may include one or more processors configured to receive a plurality of images representing an environment. The images may include image data generated by an image capture device. The processors may also be configured to transmit the image data to an image segmentation network configured to segment the images. The processors may also be configured to receive sensor data associated with the environment including sensor data generated by a sensor of a type different than an image capture device. The processors may be configured to associate the sensor data with segmented images to create a training dataset. The processors may be configured to transmit the training dataset to a machine learning network configured to run a sensor data segmentation model, and train the sensor data segmentation model using the training dataset, such that the sensor data segmentation model is configured to segment sensor data.

Abstract: A method for environmental representation, in which two images of an environment (U) are taken respectively and a disparity image is determined by means of stereo image processing. An unobstructed free space (F) is identified in the disparity image, in that each pixel of the disparity image is allocated either to the unobstructed ground surface (B) or to one of several segments (S11 to Snu) depending on disparity values of the respective pixel. Segments (S11 to Snu) of the same width are formed from pixels of the same or similar distance to an image plane. An object (O1 to Ok) located outside of the free space (F) is modelled in the environment (U) using one segment (S11 to Snu) or several segments (S11 to Snu).

Abstract: A method for environmental representation, in which two images of an environment (U) are taken respectively and a disparity image is determined by means of stereo image processing. An unobstructed free space (F) is identified in the disparity image, in that each pixel of the disparity image is allocated either to the unobstructed ground surface (B) or to one of several segments (S11 to Snu) depending on disparity values of the respective pixel. Segments (S11 to Snu) of the same width are formed from pixels of the same or similar distance to an image plane. An object (O1 to Ok) located outside of the free space (F) is modelled in the environment (U) using one segment (S11 to Snu) or several segments (S11 to Snu).

Abstract: A fishing rod that includes a rod portion having an outer circumference, at least one line guide portion, a reel secured to the rod portion, a line that extends through the line guide portion and has a hook secured to the end thereof, a handle portion that extends at least partially around the outer diameter of the rod portion, and a hook receptacle defined in the handle portion. The hook can be removably secured in the hook receptacle.

Abstract: A method of treating recurrent aphthous stomatitis in a human in need thereof. The method comprises administering a tetracycline compound in an amount that is effective to treat recurrent aphthous stomatitis, but has substantially no antibacterial activity, without administering a bisphosphonate compound.

Abstract: The battery charger provides a current to a battery depending on its type. The type of the battery is determined using, a two-pronged fork and a pusher slider assembly which may actuate two switches using a spring leaf. The two switches enable to provide a current depending on the type of battery. AAA-type, AA-type, C-type and D-type batteries may be used. The battery charger also enables to easily position the battery in the battery charger.

Abstract: The battery charger provides a current to a battery depending on its type. The type of the battery is determined using, in the preferred embodiment of the invention, a two-pronged fork and a pusher slider assembly which may actuate two switches using a spring leaf. The two switches enable to provide a current depending on the type of battery. In the preferred embodiment of the invention, AAA-type, M-type, C-type and D-type batteries may be used. The invention also enables to easily position the battery in the battery charger.

Abstract: A speech synthesizer is disclosed in which instantaneous conversational speech can be produced by an operator. The speech synthesizer comprises a two dimensional input device, such as a joystick or a playing tablet, for producing vowel-like sounds, a plurality of selection keys for producing consonant-like sounds and a third control for varying the pitch or inflection of the produced signal. The electronic circuit for producing the voicing wave forms can be either analog or digital. The system features simultaneous and continuous control of two formants.