Abstract: Techniques for determining labels for potentially small, moving, and non-impeding objects in an environment are disclosed. Unlabeled lidar segments may be evaluated to determine whether segments are associated with objects located in a drivable region and to determine one or more characteristics of such segments. This determination includes comparing segment characteristics to various criteria that may be associated with small, dynamic objects, such as object size, motion, occlusion, and/or solidity. Based on this evaluation and whether the segment is in a drivable region, the system determines whether to label the segment as a small, dynamic, non-impeding object.

Abstract: Techniques are discussed herein for controlling autonomous vehicles within a driving environment, including generating and using bounding contours associated with objects detected in the environment. Image data may be captured and analyzed to identify and/or classify objects within the environment. Image-based and/or lidar-based techniques may be used to determine depth data associated with the objects, and a bounding contour may be determined based on the object boundaries and associated depth data. An autonomous vehicle may use the bounding contours of objects within the environment to classify the objects, predict the positions, poses, and trajectories of the objects, and determine trajectories and perform other vehicle control actions while safely navigating the environment.

Abstract: Techniques for determining a probability that an object that has been appropriately labeled as a non-impeding object are disclosed. Non-impeding objects may be objects in an environment that do not substantially affect the operation of a vehicle in the environment, such as steam, fog, smoke, etc. A machine-learned model may classify such objects as non-impeding objects. The designation may be checked by using the lidar data and measurements associated with the lidar points associated with the object to determine a probability that the object is a non-impeding object. The lidar data may be aggregated within the individual voxels of a voxelized representation of the object for use in the probability determination.

Abstract: Techniques for detecting and classifying objects using lidar data are discussed herein. In some cases, the system may be configured to utilize a predetermined number of prior frames of lidar data to assist with detecting and classifying objects. In some implementations, the system may utilize a subset of the data associated with the prior lidar frames together with the full set of data associated with a current frame to detect and classify the objects.

Abstract: Techniques for identifying lidar points associated with static objects, and using such lidar points to annotate objects within two-dimensional images are discussed herein. In some examples, an object manager may receive accumulations of lidar data captured from lidar devices of a vehicle while traversing within a driving environment. In some examples, the object manager may receive a plurality of annotated images. Such annotations may identify static objects within the driving environment. In some instances, the object manager may project a lidar point into an annotated image and determine that the lidar point is associated with an annotated pixel. Based on the pixel being associated with the annotated object, the object manager may determine that the lidar point is associated with object. In some examples, the object manager may determine a subset of lidar points that are associated with the object.

Abstract: In some implementations, a classification system may receive credentials associated with a data source and may receive, from the data source and using the credentials, a set of structured data including input events and output events. The classification system may filter the set of structured data by applying a first set of rules to generate a filtered set of structured data and may convert the filtered set of structured data to one or more numerical vectors, where a vector space associated with the one or more numerical vectors is infinite-dimensional. The classification system may further cluster the one or more numerical vectors using a first machine learning model to generate one or more clusters. Accordingly, the classification system may determine one or more classifications based on the set of structured data, each of the one or more classifications being associated with a corresponding frequency and a corresponding category.

Abstract: In some implementations, a classification system may receive credentials associated with a data source and may receive, from the data source and using the credentials, a set of structured data including input events and output events. The classification system may filter the set of structured data by applying a first set of rules to generate a filtered set of structured data and may convert the filtered set of structured data to one or more numerical vectors, where a vector space associated with the one or more numerical vectors is infinite-dimensional. The classification system may further cluster the one or more numerical vectors using a first machine learning model to generate one or more clusters. Accordingly, the classification system may determine one or more classifications based on the set of structured data, each of the one or more classifications being associated with a corresponding frequency and a corresponding category.

Abstract: Techniques for densifying annotations associated with sensor data are discussed herein. For example, techniques may include projecting a subset of lidar points into an unannotated two-dimensional image. Based on the projecting, the annotations of lidar points projected into the image can be associated with the image data. A subset of pixels of pixels can be dilated, and the dilated pixels can be determined to intersect (e.g., overlap, touch, etc.). Based on the pixels intersecting (or being adjacent) and being associated with the same segment identifier, a contour can be rendered around the corresponding non-dilated pixels to identify an object in the image data.

Abstract: Techniques for identifying lidar points associated with static objects, and using such lidar points to annotate objects within two-dimensional images are discussed herein. In some examples, an object manager may receive accumulations of lidar data captured from lidar devices of a vehicle while traversing within a driving environment. In some examples, the object manager may receive a plurality of annotated images. Such annotations may identify static objects within the driving environment. In some instances, the object manager may project a lidar point into an annotated image and determine that the lidar point is associated with an annotated pixel. Based on the pixel being associated with the annotated object, the object manager may determine that the lidar point is associated with object. In some examples, the object manager may determine a subset of lidar points that are associated with the object.

Abstract: Techniques for detecting and classifying objects using lidar data are discussed herein. In some cases, the system may be configured to utilize a predetermined number of prior frames of lidar data to assist with detecting and classifying objects. In some implementations, the system may utilize a subset of the data associated with the prior lidar frames together with the full set of data associated with a current frame to detect and classify the objects.

Abstract: A vehicle can include various sensors to detect objects in an environment. In some cases, the object may be within a planned path of travel of the vehicle. In these cases, leaving the planned path may be dangerous to the passengers so the vehicle may, based on dimensions of the object, dimensions of the vehicle, and semantic information of the object, determine operational parameters associate with passing the object while maintaining a position within the planned path, if possible.

Abstract: Systems and methods disclose herein procedures for accelerated tree learning. In one class, the acceleration is based on self-adapting learning rates, while in another class, the acceleration is based on a plurality of learning rates, wherein each learning rate varies over the training; each learning rate increases linearly as a respective pseudo residual maintains a direction across sequential training iterations; and each learning rate decreases exponentially as the respective pseudo residual changes direction across sequential training iterations. The latter can be incorporated with other methodologies, such as momentum-augmented gradient boosting and Nesterov Accelerated Gradient Boosting. These systems and methods for accelerated tree learning exhibit a marked reduction in training time and resources required for gradient boosted trees.

Abstract: A vehicle can include various sensors to detect objects in an environment. In some cases, the object may be within a planned path of travel of the vehicle. In these cases, leaving the planned path may be dangerous to the passengers so the vehicle may, based on dimensions of the object, dimensions of the vehicle, and semantic information of the object, determine operational parameters associate with passing the object while maintaining a position within the planned path, if possible.

Abstract: Techniques for determining ground surface voxels based on multiresolution planes are discussed herein. Such multiresolution planes may be generated, validated, and used to connect voxels associated with the ground. In some examples, a vehicle may use capture or otherwise receive lidar data while traversing within a driving environment. The vehicle may associate the lidar data with a voxel space. Voxels can be associated with a cell of a multiresolution grid and the vehicle may determine multiresolution planes associated with the multiresolution space. For example, the vehicle may determine a first plane associated with a first resolution level, in addition to determining a plurality of planes associated with a second resolution level. The vehicle may determine the validity of the plane(s) at the various resolution levels. Based on the validity of the planes, the vehicle may connect the valid planes that are associated with the ground surface.

Abstract: Techniques are discussed herein for controlling autonomous vehicles within a driving environment, including generating and using bounding contours associated with objects detected in the environment. Image data may be captured and analyzed to identify and/or classify objects within the environment. Image-based and/or lidar-based techniques may be used to determine depth data associated with the objects, and a bounding contour may be determined based on the object boundaries and associated depth data. An autonomous vehicle may use the bounding contours of objects within the environment to classify the objects, predict the positions, poses, and trajectories of the objects, and determine trajectories and perform other vehicle control actions while safely navigating the environment.

Abstract: Various embodiments may relate to an optical device. The optical device may include a stacked structure having a first surface and a second surface opposite the first surface. The stacked structure may include a plurality of holes or grooves extending from the first surface towards the second surface. The stacked structure may include a transition metal dichalcogenide material (TMDC) material. A thickness of the stacked structure may be of any value less than 100 nm.

Abstract: Techniques for collision avoidance using an object contour are discussed. A trajectory associated with a vehicle may be received. Sensor data can be received from a sensor associated with the vehicle. A bounding contour may be determined and associated with an object represented in the sensor data. Based on the trajectory, a simulated position of the vehicle can be determined. Additionally, a predicted position of the bounding contour can be determined. Based on the simulated position of the vehicle and the predicted position of the bounding contour, a distance between the vehicle and the object may be determined. An action can be performed based on the distance between the vehicle and the object.

Abstract: Techniques are discussed herein for controlling autonomous vehicles within a driving environment, including generating and using bounding contours associated with objects detected in the environment. Image data may be captured and analyzed to identify and/or classify objects within the environment. Image-based and/or lidar-based techniques may be used to determine depth data associated with the objects, and a bounding contour may be determined based on the object boundaries and associated depth data. An autonomous vehicle may use the bounding contours of objects within the environment to classify the objects, predict the positions, poses, and trajectories of the objects, and determine trajectories and perform other vehicle control actions while safely navigating the environment.

Abstract: Techniques for determining an object contour are discussed. Depth data associated with an object may be received. The depth data, such as lidar data, can be projected onto a two-dimensional plane. A first convex hull may be determined based on the projected lidar data. The first convex hull may include a plurality of boundary edges. A longest boundary edge, having a first endpoint and a second endpoint, can be determined. An angle can be determined based on the first endpoint, the second endpoint, and an interior point in the interior of the first convex hull. The longest boundary edge may be replaced with a first segment based on the first endpoint and the interior point, and a second segment based on the interior point and the second endpoint. An updated convex hull can be determined based on the first segment and the second segment.

Abstract: A semiconductor wafer (1a, 1b) including a plurality of chips (2) and a separation zone (3) spacing the semiconductor chips (2) from each other in this wafer (1a, 1b), such a separation zone (3) extending from a front face (4a) to an opposite backside face (4b) of this wafer (1a, 1b), this separation zone (3) includes a scribe line (6) configured to be diced using plasma etching and an inlet area (13) of this scribe line (6), the inlet (13) being delimitated by free ends of plasma etch-resistant material layers (9) extending each from a peripheral wall (20) of a functional part (18) of a chip (2) into the scribe line (6) by overlapping a top of a seal ring (7) of this chip (2).