Abstract: A method for range detection is described. The method includes segmenting an image into one or more segmentation blobs captured by a monocular camera of an ego vehicle. The method includes focusing on pixels forming a selected segmentation blob of the one or more segment blobs. The method also includes determining a distance to the selected segmentation blob according to a focus function value of the monocular camera of the ego vehicle.

Abstract: Systems and methods are provided for controlling one or more brakes of the vehicle. The system may include a processor and a memory in communication with the processor with a brake control module. The brake control module includes instructions that, when executed by the processor, cause the processor to control the one or more brakes of the vehicle when the vehicle is in a first mode using a brake-by-wire system. When the vehicle is in a second mode, control the brake-by-wire system such that the one or more brakes of the vehicle are controlled using a mechanical braking system and the brake-by-wire system.

Abstract: Systems and methods of trajectory planning for an autonomous vehicle are disclosed. Exemplary implementations may: determine a first trajectory plan for the vehicle traveling along a first spatial location at a first point in time, the first trajectory plan being a reference trajectory plan; compute an optimal sequence for the vehicle traveling along a second spatial location at a second point in time subsequent the first point in time; and calculate a second trajectory plan for the vehicle by updating the first trajectory plan with information from the computed optimal sequence.

Abstract: System, methods, and other embodiments described herein relate to steering a vehicle based during loss of traction. In one arrangement, a method for steering a vehicle during loss of traction is disclosed. The method includes, responsive to detecting a slipping tire of a vehicle losing traction with a road, automatically steering the vehicle separately from an input of a steering wheel of the vehicle to cause the vehicle to follow a path. The method also includes decoupling control of a pair of front tires of the vehicle by the steering wheel. The method further includes rotating, independently of an input to the steering wheel and in parallel with steering the vehicle, the steering wheel to match an actual yaw of the vehicle.

Abstract: A method for labeling a data set by a coding model includes generating multiple sets of related initial labels based on processing a data set with a group of initial labels. The method also includes determining a quantity of occurrences, within the data set, of each one of the group of initial labels and each related initial label of the multiple sets of related initial labels. The method further includes determining, for each initial label of the group of initial labels, a breadth score based on the number of occurrences of each related initial label. The method still further includes updating one or more of the group of initial labels based on respective breadth scores satisfying a label updating condition. The method also includes labeling the data set based on the group of initial labels and the multiple sets of related initial labels.

Abstract: System, methods, and other embodiments described herein relate to steering a vehicle based on an intended yaw rate during loss of traction. In one arrangement, a method for steering a vehicle during loss of traction is disclosed. The method includes detecting a slipping tire of a vehicle losing traction with a road. The method also includes decoupling control of a pair of front tires of the vehicle by a steering wheel of the vehicle. The method also includes identifying an intended yaw rate for controlling the vehicle by detecting an angle of the steering wheel while the slipping tire does not have traction. The method further includes steering the vehicle separately from an input of a steering wheel of the vehicle to cause intended yaw rate.

Abstract: A method for controlling a vehicle in an environment includes generating, via a cross-attention model, a cross-attention cost volume based on a current image of the environment and a previous image of the environment in a sequence of images. The method also includes generating combined features by combining cost volume features of the cross-attention cost volume with single-frame features associated with the current image. The single-frame features may be generated via a single-frame encoding model. The method further includes generating a depth estimate of the current image based on the combined features. The method still further includes controlling an action of the vehicle based on the depth estimate.

Abstract: Systems and methods for a powered, robotic exoskeleton, or exosuit, for a user's limbs and body are provided. The exosuit may be equipped with airbag devices mounted at various locations on the suit. The exosuit may include on-board computing equipment that can sense, compute control commands in real-time, and actuate limbs and airbags to restore stability (fall prevention) and minimize injuries due to falls, should they happen (fall protection).

Abstract: Deformable sensors and methods for magnetically modifying membrane stiffness are provided. A deformable sensor may include a membrane coupled to a housing to form a sensor cavity. The deformable sensor may further include a first magnet located on an inner surface of the membrane in the sensor cavity. The deformable sensor may additionally include a magnetic object located at a base within the sensor cavity. The magnetic object may be configured to modifiably repel the first magnet and modify stiffness of the deformable sensor based upon the modified repulsion.

Abstract: A method for controlling a robotic device includes observing a first object associated with an object type at one or more first locations in an environment over a period of time prior to a current time. The method also includes generating a probability distribution associated with the one or more first locations based on observing the first object over the period of time. The method further includes observing, at the current time, a second object associated with the object type at a second location in the environment. The method still further includes determining a probability of the second object being at the second location based on observing the second object at the second location. The probability is based on the probability distribution associated with the one or more first locations. The method also includes controlling the robotic device to perform an action based on the probability being less than a threshold.

Abstract: A method for 3D object tracking is described. The method includes inferring first 2D semantic keypoints of a 3D object within a sparsely annotated video stream. The method also includes matching the first 2D semantic keypoints of a current frame with second 2D semantic keypoints in a next frame of the sparsely annotated video stream using embedded descriptors within the current frame and the next frame. The method further includes warping the first 2D semantic keypoints to the second 2D semantic keypoints to form warped 2D semantic keypoints in the next frame. The method also includes labeling a 3D bounding box in the next frame according to the warped 2D semantic keypoints in the next frame.

Abstract: A method for ground state inference is described. The method includes modeling a material state of a selected material. The method also includes inferring an energy function and a ground state of the selected material according to the modeling of the material state. The method further includes predicting a different material state of the selected material in response to the inferring of the ground state of the material.

Abstract: Systems and methods for detecting roadway lane boundaries are disclosed herein. One embodiment receives image data of a portion of a roadway; receives historical vehicle trajectory data for the portion of the roadway; generates, from the historical vehicle trajectory data, a heatmap indicating, for a given pixel in the heatmap, an extent to which the given pixel coincides spatially with vehicle trajectories in the historical vehicle trajectory data; and projects the heatmap onto the image data to generate a composite image that is used in training a neural network to detect roadway lane boundaries, the projected heatmap acting as supervisory data. The trained neural network is deployed in a vehicle to generate and save map data including detected roadway lane boundaries for use by other vehicles or to control operation of the vehicle itself based, at least in part, on roadway lane boundaries detected by the trained neural network.

Abstract: Systems and methods to improve machine learning by explicitly over-fitting environmental data obtained by an imaging system, such as a monocular camera are disclosed. The system includes training self-supervised depth and pose networks in monocular visual data collected from a certain area over multiple passes. Pose and depth networks may be trained by extracting data from multiple images of a single environment or trajectory, allowing the system to overfit the image data.

Abstract: Systems and methods are provided for influential control over a driver's operation of a vehicle's steering wheel. Upon issuing an autonomous control signal to control motive operation of the vehicle, an autonomous control system of the vehicle may reinforce or influence the application of the autonomous control signal by inducing or cuing to induce the driver's hand(s) to turn the vehicle's steering wheel in a particular direction or by a particular amount/with a particular level of torque.

Abstract: Systems, methods, and other embodiments described herein relate to downsampling training data so as to simplify the training of models as well as increase the prediction accuracy of the models. In one embodiment, a method includes training a model on a dataset to learn a covariance function, determining a covariance between a selected data value and the dataset using the covariance function, selecting a subset from the dataset based on the covariance, and predicting one or more potential experiments based on the subset.

Abstract: A method includes receiving data relating to pedestrian activity at one or more locations outside of a crosswalk, analyzing the data, based on the data, identifying at least one location of the one or more locations as a constructive crosswalk, and controlling operation of an autonomous vehicle based on the at least one location of the constructive crosswalk.

Abstract: A vehicle exhaust dissipation system includes a fan unit and one or more processors. A memory is communicably coupled to the one or more processors and stores a fan control module including computer-readable instructions that when executed by the processor(s) cause the processor(s) to, responsive to a determination that an ambient temperature is equal to or below a fan rotor activation temperature, control operation of the fan unit to generate an airflow directed into an exhaust gas exiting the vehicle from a vehicle exhaust opening.

Abstract: A method for multitask learning based on Hermitian operators is described. The method includes training a multitask machine learning (MTML) model to map an input representation of a material onto a complex wave function state vector. The method also includes inferring, by a trained, MTML model, observable property matrices for each observable property of the material. The method further includes converting the observable property matrices into complex Hermitian operators. The method also includes predicting target properties of the material according to the complex Hermitian operators and the complex wave function state vector.

Abstract: A method for learning depth-aware keypoints and associated descriptors from monocular video for ego-motion estimation is described. The method includes training a keypoint network and a depth network to learn depth-aware keypoints and the associated descriptors. The training is based on a target image and a context image from successive images of the monocular video. The method also includes lifting 2D keypoints from the target image to learn 3D keypoints based on a learned depth map from the depth network. The method further includes estimating ego-motion from the target image to the context image based on the learned 3D keypoints.