Abstract: Systems and methods of controlling a clutch in a hybrid vehicle with a manual transmission, are provided. With the goal of enabling autonomous/assisted clutch control in a hybrid vehicle, while preserving the familiar mechanical feeling at the clutch pedal that driving enthusiasts prefer, embodiments of the disclosed technology use a shuttle valve to blend control of clutch engagement between a driver and an ECU. In these embodiments, a clutch pedal in the vehicle may be mechanically connected to a piston in a first hydraulic cylinder (just like in a traditional mechanical/hydraulic clutch actuation system), and an ECU may actuate a second hydraulic cylinder. Accordingly, a shuttle valve may be used to route the fluid coming from the cylinder with the greater pressure (i.e. the driver actuated cylinder or the ECU actuated cylinder), to a third hydraulic cylinder which adjusts engagement of a clutch by a mechanical linkage.

Abstract: A method for 3D object detection is described. The method includes predicting, using a trained monocular depth network, an estimated monocular input depth map of a monocular image of a video stream and an estimated depth uncertainty map associated with the estimated monocular input depth map. The method also includes feeding back a depth uncertainty regression loss associated with the estimated monocular input depth map during training of the trained monocular depth network to update the estimated monocular input depth map. The method further includes detecting 3D objects from a 3D point cloud computed from the estimated monocular input depth map based on seed positions selected from the 3D point cloud and the estimated depth uncertainty map. The method also includes selecting 3D bounding boxes of the 3D objects detected from the 3D point cloud based on the seed positions and an aggregated depth uncertainty.

Abstract: A method for generating a dense light detection and ranging (LiDAR) representation by a vision system of a vehicle includes generating, at a depth estimation network, a depth estimate of an environment depicted in an image captured by an image capturing sensor integrated with the vehicle. The method also includes generating, via a sparse depth network, one or more sparse depth estimates of the environment, each sparse depth estimate associated with a respective sparse representation of one or more sparse representations. The method further includes generating the dense LiDAR representation based on a dense depth estimate that is generated based on the depth estimate and the one or more sparse depth estimates. The method still further includes controlling an action of the vehicle based on the dense LiDAR representation.

Abstract: A method for adjusting a direction of focus for a driver includes identifying a first direction of focus, of the driver of a vehicle, at a first time period in accordance with monitoring one or both of a gaze direction of the driver or a head pose of the driver. The method further includes generating audio from one or more speakers of a set of speakers located within a cabin of the vehicle based on identifying the first direction of focus, a spatial location of the audio corresponding to an expected direction of focus. The method also includes terminating the audio signal in accordance with a second direction of focus, at a second time period, being within a range of the expected direction of focus, the second direction of focus being determined based on the monitoring of one or both of the gaze direction or the head pose.

Abstract: A method for controlling the acceleration rate of a vehicle based on the vehicle's location includes determining, via one or more sensors, that the vehicle is at a first location and reducing the vehicle's initial target acceleration rate to a first adjusted target acceleration rate in response to the location satisfying a first location condition. Subsequently, the method includes determining that the vehicle has reached a second location and increasing the first adjusted target acceleration rate to a second adjusted target acceleration rate in accordance with determining that the vehicle has reached the second location. This method can be applied to vehicles operating in autonomous or semi-autonomous modes.

Abstract: A method for training a driver to operate a vehicle includes identifying a first head pose direction of a driver, at a first time period, in accordance with monitoring the driver. The method also includes activating one or more first magnets of a first set of magnets to interact with one or more second magnets of a second set of magnets in accordance with a difference between the first head pose direction of the driver and an expected head pose direction. The first set of magnets may be integrated with a headrest of the vehicle, and the second set of magnets may be integrated with a helmet worn by the driver. The method further includes deactivating the one or more first magnets in accordance with a second head pose direction of the driver, at a second time period, being within a range of the expected head pose direction.

Abstract: A method for training a driver to navigate a path includes displaying a virtual vehicle on one or more display units integrated with a vehicle, the virtual vehicle being displayed as following the vehicle. The method also includes monitoring a location of the vehicle and a speed of the vehicle along the path. The method further includes adjusting one or both of an angle of the virtual vehicle in relation to the vehicle or a distance between the virtual vehicle and the vehicle based on monitoring the location and the speed, such that the driver adjusts one or both of a lateral position of the vehicle on the path or the speed of the vehicle in accordance with one or both of the adjusted angle or the adjusted distance.

Abstract: A system includes a processor configured to build a model of human behavior that can be assigned to a simulated human, where the model is trained on a dataset of interactions between humans, and parameters of the model can be adjusted. The processor is further configured to build a model of a task to be engaged in by a plurality of simulated humans, simulate interactions between the plurality of simulated humans, and display behavior of the simulated humans and information about the interactions between the simulated humans during simulated interactions.

Abstract: Systems and methods for automatically generating solver code for a nonlinear model predictive controller are disclosed. In one embodiment, a method of automatically generating solver code for a nonlinear model predictive control solver includes receiving an optimal control problem code, wherein the optimal control problem code represents an optimal control problem comprising a cost function, one or more constraints, and a continuous time model representing dynamics of a system. The method further includes receiving a discretization method preference, a linearization point preference, and a parameter specification, and encoding the optimal control problem into an optimization problem by discretizing the optimal control problem according to the discretization method preference, and linearizing the optimal control problem according to the linearization point preference. The method further includes generating the solver code from the optimization problem.

Abstract: Lifting robot systems and methods for operating the same are disclosed. A lifting robot system includes a sensor device and a robot device. The robot device includes a body, a lifting component movably coupled to the body, and a collection tray coupled to the lifting component. Upon receiving a command to lift an object, the sensor device automatically detects the object, the robot device places the object on the collection tray, and the robot device causes the lifting component to lift the collection tray from a first height to a second height.

Abstract: A method for a physical design tool to recommend design alternatives based on responses to semantic prompts is described. The method includes identifying first design action patterns elicited by a specific semantic prompt that differ across individual designers based on historical data. The method also includes generating sequences of actions with varying similarity to the individual designers to present alternatives to new designers. The method further includes identifying second design action patterns that differ across sematic prompts with different linguistic properties. The method also includes training a behavioral model of each of the individual designers based on responses to the specific semantic prompt, responses to the sematic prompts with the different linguistic properties, and the sequences of actions with the varying similarity. The method further includes displaying the design alternatives recommended to an individual designer based on the behavioral model created for the individual designer.

Abstract: Systems and methods are provided for optimizing a driver's trajectory around a race track. The system can receive driver data representing driver inputs to a vehicle. A vehicle state can be determined for at least one location on the race track based on the received driver data. Expert driver data can be determined representing inputs of an expert driver at the determined vehicle state for the at least one location. The system can predict how the expert driver would proceed from each current vehicle state based on the expert driver data and map the current vehicle state at the at least one location to a recommended trajectory from each current vehicle state based on predicting how the expert driver would proceed.

Abstract: An electric vehicle (EV) battery pack replacement system includes a replacement pack ramp structure having a replacement pack ramp, a vehicle-mountable battery pack receptacle, and a vehicle-mountable replacement pack coupling system structured to be operable to connect a vehicle to a replacement pack positioned on the ramp during movement of the vehicle over the ramp. Further movement of the vehicle over the ramp after connection to the replacement pack to the vehicle moves the replacement pack up the ramp. The ramp is structured so that movement of the replacement pack up the ramp causes the replacement pack to be positioned in in the receptacle. The system uses continuous motion of the vehicle and the replacement pack coupling system for loading a new battery, obviating the need for robotics or other active or powered infrastructure external to the vehicle.

Abstract: One aspect of the disclosure relates to systems and methods for determining probabilities of successful synthesis of materials in the real world at one or more points in time. The probabilities of successful synthesis of materials in the real world at one or more points in time can be determined by representing the materials and their pre-defined relationships respectively as nodes and edges in a network form, and computation of the parameters of the nodes in the network as input to a classification model for successful synthesis. The classification model being configured to determine probabilities of successful synthesis of materials in the real world at one or more points in time.

Abstract: Systems and methods are provided for evaluating different vehicle and race configurations in simulation through deriving a driving model that can accurately account for driver preferences, environment settings and changes in vehicle setup without using independent models. The driving model may be used to simulate race scenarios to determine optimal vehicle setup parameters, optimal goals and optimal functions to be implemented on a vehicle during a race. The systems and methods may include generating goals according to training data; determining functions to achieve the goals based on algorithms; training a driving model based on the training data, the goals and the functions; performing one or more simulations of a set of goals and a set of functions derived from the driving model based on first race characteristics; and generating results comprising an optimal vehicle setup, optimal goals and optimal functions according to the simulations.

Abstract: Systems and methods for providing updatable roadway codes are described. One embodiment of a method includes locating a roadway code along a roadway and determining stored data provided by the roadway code. Some embodiments include providing at least a portion of the stored data to the user and capturing environmental data along the roadway, where the environmental data corresponds with the stored data. Still some embodiments include comparing the stored data with the environmental data to determine whether the stored data is accurate and in response to determining that the stored data is not accurate, providing a communication to update the stored data.

Abstract: Extended reality content in a video can be varied based on driver attentiveness. The video can be of an external environment of a vehicle and can be presented in real-time on a display located within the vehicle. The display can be a video pass through display. The display can be an in-vehicle display, or it can be a part of a video pass-through extended reality headset. The video can present a view of an external environment of the vehicle as well as extended reality content. A level of attentiveness of a driver of the vehicle can be determined. An amount of the extended reality content presented in the video can be varied based on the level of attentiveness.

Abstract: Deformable sensors having a stereo depth sensor and an infrared sensor and methods of their use are disclosed. In one embodiment, a deformable sensor includes an enclosure having a housing and a deformable membrane, wherein the deformable membrane is transparent to a wavelength band in the infrared spectrum, a stereo depth sensor that is disposed within the enclosure and is configured to view an underside of the deformable membrane, and output a deformation region of the deformable membrane as a result of contact with an object, wherein the deformation region includes depth information, and an infrared sensor that is disposed within the enclosure and is configured to view the object through the deformable membrane when the object contacts the deformable membrane and output a contact patch region corresponding with a region of the deformable membrane that contacts the object.

Abstract: An embodiment of the present disclosure takes the form of a sensor device including a base structure with a plurality of ports extending into the base structure, a plurality of cells within the base structure and fluidly coupled to the plurality of ports, and an electrically conductive material disposed within the cells. The sensor device also includes a sensorized lattice structure disposed on the base structure and an electrically conductive layer disposed on top of the sensorized lattice structure.

Abstract: Systems, methods, and other embodiments described herein relate to encouraging exploration within a computing application. In one embodiment, a system includes a processor and a memory storing machine-readable instructions. The instructions, when executed by the processor, cause the processor to identify when a user is stuck in a repetitive pattern within a computing application. When the user is stuck in the repetitive pattern the instructions, the instructions when executed by the processor, cause the processor to encourage the user to further explore the computing application by 1) intervening in user activity within the computing application and 2) presenting attention-capturing content that is targeted to the user and unrelated to a domain of content accessed by the user within the computing application.