Abstract: Various systems and methods for implementing LiDAR-based object tracking described herein. An object tracking system for a vehicle includes an interface to communicate with object detection circuitry and an object similarity calculator circuitry, to obtain segmented data of an environment the object tracking system is operating within, the segmented data obtained using a light imaging detection and ranging (LiDAR) system, and the segmented data including a first plurality of segments detected in a first frame and a second plurality of segments detected in a second frame, the second frame captured after the first frame; determine, for a given segment of the first plurality of segments, a similar segment in the second plurality of segments; assign an VEHICLE object identification of the given segment to the similar segment; and track the similar segment from the first frame to the second frame based on the object identification.

Abstract: Technologies for performing sensor fusion include a compute device. The compute device includes circuitry configured to obtain detection data indicative of objects detected by each of multiple sensors of a host system. The detection data includes camera detection data indicative of a two or three dimensional image of detected objects and lidar detection data indicative of depths of detected objects. The circuitry is also configured to merge the detection data from the multiple sensors to define final bounding shapes for the objects.

Abstract: A method for motion tracking is provided including receive first data, receive second data, transform the second data to generate transformed second data corresponding to the first frame; determine a first weighting factor for the first data and a second weighting factor for the transformed second data; weight the first data using the first weighting factor to generate first weighted data; weight the transformed second data using the second weighting factor to generate second weighted data; and combine the weighted first data and the weighted second data to generate combined image data. The first data include a first frame of a first scene of an environment detected by a camera or image sensor. The second data include a second frame of a second scene of an environment detected by a light detection and ranging (LIDAR) sensor. At least a subset of the second scene corresponds to the first scene.

Abstract: Technologies for performing sensor fusion include a compute device. The compute device includes circuitry configured to obtain detection data indicative of objects detected by each of multiple sensors of a host system. The detection data includes camera detection data indicative of a two or three dimensional image of detected objects and lidar detection data indicative of depths of detected objects. The circuitry is also configured to merge the detection data from the multiple sensors to define final bounding shapes for the objects.

Abstract: Spatial data may be divided along an axis of the second dimension into a first data segment and a second data segment, such that the first data segment is limited to data points of the spatial data with second dimension coordinates within a first range and the second data segment is limited to data points of the spatial data with second dimension coordinates within a second range. A first processing element may execute an object detection process on the first data segment to generate a first list of objects within the first data segment. A second processing element may execute the object detection process on the second data segment to generate a second list of objects within the second data segment. A first set of objects detected in the first data segment may be combined with a second set of objects detected in the second data segment.

Abstract: Systems, apparatus, articles of manufacture, and methods are disclosed that direct transmission of data between network-connected devices including circuitry, instructions, and programmable circuitry to at least one of instantiate or execute the instructions to cause the interface circuitry to identify a neural network (NN) to a first device of a first combination of devices corresponding to a first network topology, cause the first device to process first data with a first portion of the NN, and cause a second device of a second combination of devices to process second data with a second portion of the NN, the second combination of devices corresponding to a second network topology.

Abstract: Various systems and methods for implementing LiDAR-based object detection and classification are described herein. An object detection system includes a feature extraction and object identification (FEOI) circuit to: receive segmented data of an environment around the object detection system, the segmented data obtained using a light imaging detection and ranging (LiDAR) system, oriented with respect to a direction of travel; compute spatial and structural parameters of a segment of the segmented data; and use the spatial and structural parameters with a machine learning model to obtain a classification of the segment.

Abstract: Embodiments of adaptive multi-modal event detection are disclosed herein. In one example, sensor data captured by multiple sensors is received, and an inconsistency among the sensors is detected based on performing event detection on the sensor data. An external environment of the sensors is detected based on the sensor data, and one or more configuration parameter(s) for event detection are adjusted based on the external environment of the sensors. Event detection is then performed on the sensor data based on the adjusted configuration parameter(s).

Abstract: Embodiments of localization and mapping using depth modeling are described herein. In one example, frames of image data captured by sensor(s) from various poses within an environment are received over an interface. Keypoints are detected in the current frame, and matching keypoints are found in preceding frames. The pose of the current frame is determined based at least partially on depth models associated with the matching keypoints.

Abstract: Methods, apparatus, systems and articles of manufacture to train a model using attestation data are disclosed. An example apparatus includes memory, instructions, and at least one processor to execute machine readable instructions to at least access training data originating from an edge device, the training data including telemetry information and attestation information, determine a weighting value to be used for the telemetry information based on the attestation information associated with the edge device, and train a machine learning model based on the telemetry information and the weighting value.

Abstract: Methods, apparatus, systems and articles of manufacture to train a model using attestation data are disclosed. An example apparatus includes a model trainer to train a machine learning model using a golden training data set received from a server to generate golden training results; and an attestation result generator to: compare the shared model training results to the golden training results; and determine if attestation of the shared model training results passes based on the comparison of the shared model training results and the golden training results.

Abstract: Spatial data may be divided along an axis of the second dimension into a first data segment and a second data segment, such that the first data segment is limited to data points of the spatial data with second dimension coordinates within a first range and the second data segment is limited to data points of the spatial data with second dimension coordinates within a second range. A first processing element may execute an object detection process on the first data segment to generate a first list of objects within the first data segment. A second processing element may execute the object detection process on the second data segment to generate a second list of objects within the second data segment. A first set of objects detected in the first data segment may be combined with a second set of objects detected in the second data segment.

Abstract: Methods, systems and apparatuses may provide for technology that conducts an automated vision analysis of image data associated with an interior of a vehicle cabin, determines a state of a child restraint system (CRS) based on the automated vision analysis, and generates an alert if the state of the CRS does not satisfy one or more safety constraints. In one example, the technology identifies the safety constraint(s) based on a geographic location of the vehicle cabin.

Abstract: Systems, apparatuses, and methods include technology that identifies a first dataset that comprises a plurality of data values, and partitions the first dataset into a plurality of bins to generate a second dataset, where the second dataset is a compressed version of the first dataset. The technology randomly subsamples data associated with the first dataset to obtain groups of randomly subsampled data, and generates a plurality of decision tree models during an unsupervised learning process based on the groups of randomly subsampled data and the second dataset.

Abstract: Spatial data may be divided along an axis of the second dimension into a first data segment and a second data segment, such that the first data segment is limited to data points of the spatial data with second dimension coordinates within a first range and the second data segment is limited to data points of the spatial data with second dimension coordinates within a second range. A first processing element may execute an object detection process on the first data segment to generate a first list of objects within the first data segment. A second processing element may execute the object detection process on the second data segment to generate a second list of objects within the second data segment. A first set of objects detected in the first data segment may be combined with a second set of objects detected in the second data segment.

Abstract: Techniques are disclosed for task priority scheduling and resource allocation of autonomous agents. The scheduling may utilize sensor data characteristics to facilitate scheduling decisions. The present disclosure also provides refinement of task priority scheduling utilizing fleet management information-based scene and environment information, such as by using information available to the fleet management controller.

Abstract: Techniques are disclosed to facilitate multi-agent path planning and to enable navigation for robotics systems. The path planning may include a Probabilistic Road Map (PRM), Deterministic Road Map (DRM) or any other kind of road maps with an adaptive sampling technique having a rapid two-way search to compute a global path. With the road map and Two-way search strategy, a parallel or simultaneous check for feasibility at each iteration may be performed. The road map can be saved for future inquiries.

Abstract: Apparatus and method to facilitate automatic detection of a device state are disclosed herein. Selectively constraining a sensor based data set associated with one or more states of a device, wherein selectively constraining the sensor based data set includes analyzing a distribution of the sensor based data set to determine whether to constrain the sensor based data set, the sensor based data set including a first class and a second class of data values. Determining a threshold associated with the sensor based data set by selecting the threshold based on a variance between the first and second classes of the sensor based data set, wherein selecting the threshold includes using a constrained sensor based data set when the sensor based data set is determined to be constrained, and wherein the threshold indicates the data values associated with the first and second classes.

Abstract: Technologies for performing sensor fusion include a compute device. The compute device includes circuitry configured to obtain detection data indicative of objects detected by each of multiple sensors of a host system. The detection data includes camera detection data indicative of a two or three dimensional image of detected objects and lidar detection data indicative of depths of detected objects. The circuitry is also configured to merge the detection data from the multiple sensors to define final bounding shapes for the objects.

Abstract: Various aspects of techniques, systems, and use cases include provide instructions for operating an autonomous mobile robot (AMR). A technique may include capturing audio or video data using a sensor of the AMR, performing a classification of the audio or video data using a trained classifier, and identifying a coordinate of an environmental map corresponding to a location of the audio or video data. The technique may include updating the environmental map to include the classification as metadata corresponding to the coordinate. The technique may include communicating the updated environmental map to an edge device.