UATC, LLC

Systems and methods for facilitating communication with autonomous vehicles are provided. In one example embodiment, a computing system can obtain a first type of sensor data (e.g., camera image data) associated with a surrounding environment of an autonomous vehicle and/or a second type of sensor data (e.g., LIDAR data) associated with the surrounding environment of the autonomous vehicle. The computing system can generate overhead image data indicative of at least a portion of the surrounding environment of the autonomous vehicle based at least in part on the first and/or second types of sensor data. The computing system can determine one or more lane boundaries within the surrounding environment of the autonomous vehicle based at least in part on the overhead image data indicative of at least the portion of the surrounding environment of the autonomous vehicle and a machine-learned lane boundary detection model.

The present disclosure is directed encoding LIDAR point cloud data. In particular, a computing system can receive point cloud data for a three-dimensional space. The computing system can generate a tree-based data structure from the point cloud data, the tree-based data structure comprising a plurality of nodes. The computing system can generate a serial representation of the tree-based data structure. The computing system can, for each respective node represented by a symbol in the serial representation: determine contextual information for the respective node, generate, using the contextual information as input to a machine-learned model, a statistical distribution associated with the respective node, and generate a compressed representation of the symbol associated with the respective node by encoding the symbol using the statistical distribution for the respective node. The computing system can generate a compressed bitstream by sequentially ordering a plurality of compressed representations associated with the plurality of symbols.

Disclosed are methods, systems, and non-transitory computer readable media that control an autonomous vehicle via at least two sensors. One aspect includes capturing an image of a scene ahead of the vehicle with a first sensor, identifying an object in the scene at a confidence level based on the image, determining the confidence level of the identifying is below a threshold, in response to the confidence level being below the threshold, directing a second sensor having a field of view smaller than the first sensor to generate a second image including a location of the identified object, further identifying the object in the scene based on the second image, controlling the vehicle based on the further identification of the object.

Systems and methods for performing semantic segmentation of three-dimensional data are provided. In one example embodiment, a computing system can be configured to obtain sensor data including three-dimensional data associated with an environment. The three-dimensional data can include a plurality of points and can be associated with one or more times. The computing system can be configured to determine data indicative of a two-dimensional voxel representation associated with the environment based at least in part on the three-dimensional data. The computing system can be configured to determine a classification for each point of the plurality of points within the three-dimensional data based at least in part on the two-dimensional voxel representation associated with the environment and a machine-learned semantic segmentation model. The computing system can be configured to initiate one or more actions based at least in part on the per-point classifications.

Generally, the disclosed systems and methods implement improved detection of objects in three-dimensional (3D) space. More particularly, an improved 3D object detection system can exploit continuous fusion of multiple sensors and/or integrated geographic prior map data to enhance effectiveness and robustness of object detection in applications such as autonomous driving. In some implementations, geographic prior data (e.g., geometric ground and/or semantic road features) can be exploited to enhance three-dimensional object detection for autonomous vehicle applications. In some implementations, object detection systems and methods can be improved based on dynamic utilization of multiple sensor modalities. More particularly, an improved 3D object detection system can exploit both LIDAR systems and cameras to perform very accurate localization of objects within three-dimensional space relative to an autonomous vehicle. For example, multi-sensor fusion can be implemented via continuous convolutions to fuse image data samples and LIDAR feature maps at different levels of resolution.

Systems and methods for answering region specific questions are provided. A method includes obtaining a regional scene question including an attribute query and a spatial region of interest for a training scene depicting a surrounding environment of a vehicle. The method includes obtaining a universal embedding for the training scene and an attribute embedding for the attribute query of the scene question. The universal embedding can identify sensory data corresponding to the training scene that can be used to answer questions concerning a number of different attributes in the training scene. The attribute embedding can identify aspects of an attribute that can be used to answer questions specific to the attribute. The method includes determining an answer embedding based on the universal embedding and the attribute embedding and determining a regional scene answer to the regional scene question based on the spatial region of interest and the answer embedding.

An autonomous vehicle computing system can include a primary perception system configured to receive a plurality of sensor data points as input generate primary perception data representing a plurality of classifiable objects and a plurality of paths representing tracked motion of the plurality of classifiable objects. The autonomous vehicle computing system can include a secondary perception system configured to receive the plurality of sensor data points as input, cluster a subset of the plurality of sensor data points of the sensor data to generate one or more sensor data point clusters representing one or more unclassifiable objects that are not classifiable by the primary perception system, and generate secondary path data representing tracked motion of the one or more unclassifiable objects. The autonomous vehicle computing system can generate fused perception data based on the primary perception data and the one or more unclassifiable objects.

Systems and methods for movement detection are provided. In one example embodiment, a computer-implemented method includes obtaining image data and range data representing a scene external to an autonomous vehicle, the image data including at least a first image and a second image that depict the scene. The method includes identifying a set of corresponding image features from the image data, the set of corresponding image features including a first feature in the first image having a correspondence with a second feature in the second image. The method includes determining a respective distance for each of the first feature and the second feature based at least in part on the range data. The method includes determining a velocity associated with a portion of a scene represented by the set of corresponding image features based at least in part on the respective distance for the first feature and the second feature.

Disclosed are methods, systems, and non-transitory computer readable media that control an autonomous vehicle via at least two sensors. One aspect includes capturing an image of a scene ahead of the vehicle with a first sensor, identifying an object in the scene at a confidence level based on the image, determining the confidence level of the identifying is below a threshold, in response to the confidence level being below the threshold, directing a second sensor having a field of view smaller than the first sensor to generate a second image including a location of the identified object, further identifying the object in the scene based on the second image, controlling the vehicle based on the further identification of the object.

Systems and methods of the present disclosure provide an improved approach for open-set instance segmentation by identifying both known and unknown instances in an environment. For example, a method can include receiving sensor point cloud input data including a plurality of three-dimensional points. The method can include determining a feature embedding and at least one of an instance embedding, class embedding, and/or background embedding for each of the plurality of three-dimensional points. The method can include determining a first subset of points associated with one or more known instances within the environment based on the class embedding and the background embedding associated with each point in the plurality of points. The method can include determining a second subset of points associated with one or more unknown instances within the environment based on the first subset of points. The method can include segmenting the input data into known and unknown instances.