Abstract: Systems, methods, and other embodiments described herein relate to determining a type and an issue time of a vehicle control command based on an estimated realization delay. In one embodiment, a method includes identifying a first realization delay associated with a first control command and identifying a second realization delay associated with a second control command. The second control command is issued after the first control command is issued. The method includes issuing the first control command at an issue time that is based on the first realization delay and the second realization delay, and realizing the first control command on a vehicle.

Abstract: Systems and methods are described for optimizing bounding boxes generated using point cloud information. In one example, a method for optimizing the bounding box includes the step of using an iterative closet point (ICP) algorithm to transform the bounding box using a transformation. The point cloud acts as the reference for the ICP algorithm, while the bounding box acts as the source for the ICP algorithm.

Abstract: A method for fitting an odometry noise model of a vehicle is described. The method includes collecting multiple passes of sensor measurements from motion sensors of the vehicle to measure a motion sensor noise during a testing operation of the vehicle. The method also includes determining an estimated standard deviation of the motion sensor noise measured during the testing operation of the vehicle. The method further includes determining the odometry noise model of the vehicle according to the estimated standard deviation of the motion sensor noise measured during the testing operation of the vehicle. The method also includes compensating, using the odometry noise model, the motion sensor noise during normal operation of the vehicle.

Abstract: Systems and methods are provided for lane biasing, e.g., positioning a vehicle within a current lane of travel. Lane biasing can be dependent on learned driver behaviors or other factors that may impact lane biasing, e.g., weather conditions, traffic conditions, and so on (which may also be learned). Lane biasing may result in a vehicle traveling, e.g., off the center-line of a current lane of travel, and is distinguished from conventional lane keep assist systems that merely position a vehicle in the center of a current lane of travel by default.

Abstract: A method performed by an autonomous vehicle includes identifying a condition preventing the autonomous vehicle from proceeding along an intended route. The method also includes prompting a passenger of the autonomous vehicle to interact with a driver of a first vehicle in response to identifying the condition. The method further includes receiving, from the passenger, an input at an interface of the autonomous vehicle indicating a successful interaction or an unsuccessful interaction with the driver. The method also includes controlling the autonomous vehicle to proceed along the intended route based on the input indicating the successful interaction with the driver.

Abstract: Systems and methods for controlling a vehicle using operational constraints, including friction estimates are disclosed. Friction estimation include estimating a tire-road coefficient of friction using bins and envelopes. Bounding envelopes are configured to ensure that stability of the vehicle is maintained. The friction estimate is used to define the bounding envelopes. Further, the bounding envelopes are received as feedback into the friction estimation, itself. Based on the bounding envelope, the friction estimation can be adjusted. Then, the adjusted friction estimation can be fed back to reshape the bounding envelopes. Multiple bins can be used to evaluate an operating range of friction. Each bin can be used to compare actual vehicle dynamics with expected dynamics based on the estimation using the range assigned to that bin. Multiple bins and multiple controllers can run in parallel to re-estimate friction considering the vehicle dynamics over time.

Abstract: A method for depth estimation performed by a depth estimation system associated with an agent includes determining a first depth of a first image and a second depth of a second image, the first image and the second image being captured by a sensor associated with the agent. The method also includes generating a first 3D image of the first image based on the first depth, a first pose associated with the sensor, and the second image. The method further includes generating a warped depth image based on transforming the first depth in accordance with the first pose. The method also includes updating the first pose based on a second pose associated with the warped depth image and the second depth, and updating the first 3D image based on the updated first pose. The method further includes controlling an action of the agent based on the updated first 3D image.

Abstract: A system is provided for implementing trajectory analysis entry controls. A trajectory analysis entry control system is implemented in a vehicle, and supports remote and automatic entry control functions for the vehicle. For example the system triggers various automatic entry controls functions, such as remotely unlocking doors and/or a trunk of a vehicle, by tracking and analyzing a user's trajectory of movement while they are approaching the vehicle. A system includes a user identification device to identify whether a user is authorized to access the vehicle, such as a driver. The system also includes a trajectory tracking device that determines a trajectory of the user with respect to the vehicle. As the user's trajectory is determined to be approaching the vehicle, the trajectory tracking device further determines which entry point of the vehicle the user is approaching and automatically engages an automatic unlocking of that access point of the vehicle.

Abstract: A system for producing a trajectory from a diverse set of possible movements can include a processor and a memory. The memory can store a possible movement production module and a trajectory production module. The possible movement production module can include instructions that cause the processor to: (1) produce a set of possible movements of a mobile robot with respect to a challenge, (2) determine a measure of diversity between a first possible movement and a second possible movement, and (3) produce a subset. Measures of diversity associated with possible movements included in the subset can be larger than measures of diversity associated with possible movements excluded from the subset. The trajectory production module can include instructions that cause the processor to: (1) produce trajectories associated with the possible movements included in the subset and (2) cause the mobile robot to move according to one of the trajectories.

Abstract: Systems and methods are provided for variable actuation and release of a vehicle's brake hold mechanism/function. A brake hold device may be implemented as a “tuner” or controller that can connect to and override, e.g., the default functionality of an existing brake hold mechanism of a vehicle. Delays in exiting a brake hold state, repetitive actuation of an accelerator to exit a brake hold state, and other shortcomings associated with the conventional operation of brake hold mechanisms may be avoided by taking into account relevant vehicle operating conditions or environmental/road/traffic conditions.

Abstract: The disclosure generally describes a system for monitoring a storage compartment of a vehicle comprising a first sensor configured to detect a loading or unloading event, a second sensor configured to capture data regarding one or more items within the storage compartment of the vehicle, wherein the second sensor captures video data of the one or more items within the storage compartment of the vehicle in response to the first sensor detecting a loading or unloading event, and a virtual imaging system configured to classify each item in the storage compartment, generate a virtual replication of each item within the storage compartment, and transmit the virtual replication to a user device.

Abstract: System, methods, and other embodiments described herein relate to improving the detection of traffic lights associated with a driving lane using a camera instead of map data. In one embodiment, a method includes estimating, from an image using a first model, depth and orientation information of traffic lights relative to a driving lane of a vehicle. The method also includes computing, using a second model, relevancy scores for the traffic lights according to geometric inferences between the depth and the orientation information. The method also includes assigning, using the second model, a primary relevancy score for a light of the traffic lights associated with the driving lane according to the depth and the orientation information. The method also includes executing a control task by the vehicle according to the primary relevancy score and a state confidence, computed by the first model, for the light.

Abstract: Systems and methods for optimizing the loading or unloading of cargo in a cargo area is provided. Friction-reducing elements, such as rails, may be disposed atop a surface of the cargo area. The friction-reducing elements vibrate pursuant to the application of vibrational energy generated by one or more vibration actuators operatively connected, directly or indirectly, to the friction-reducing elements. The application of vibrational energy reduces the time that cargo contacts the friction-reducing elements, which in turn, reduces the amount of friction applied to the cargo, easing the burden of loading/unloading the cargo.

Abstract: In accordance with one embodiment of the present disclosure, a sensor system includes a sensing surface and an array of pressure sensors arranged on the sensing surface. The array of pressure sensors includes at least one pressure sensor is parallel to the sensing surface, at least one pressure sensor is angled between parallel and perpendicular to the sensing surface, and at least one pressure sensor is perpendicular to the sensing surface. The pressure sensors are micro electro mechanical system (MEMS) barometric pressure sensors.

Abstract: A method for controlling a comfort component of a vehicle includes setting a comfort component, while a passenger is occupying the vehicle, to an active state to adjust a temperature within the vehicle to a passenger desired temperature based on receiving an input from the passenger. The method further includes maintaining, after the passenger exited the vehicle, the comfort component in an active state based on a likelihood of the passenger returning to the vehicle satisfying an active state condition. The method still further includes capturing, after the passenger exited the vehicle, an image of a person approaching the vehicle via one or more sensors of the vehicle. The method also includes adjusting the comfort component from the active state to a default state based on an identity of the person being different from an identity of the passenger, the identity of the person being determined based on the image.

Abstract: The disclosure generally relates to methods for assisting a driver of a vehicle that include monitoring the driver’s actions, during a first portion of time that starts when the driver begins to engage in a surveillance type action and ends when the driver ends the surveillance type action, activating one or more feedback devices, when the driver engages in a surveillance type action, monitoring the driver’s actions during a second portion of time that beings when the driver ends the surveillance type action, and ends when the driver begins the surveillance type action, and activating one or more feedback devices, during the second portion of time when the driver engages in a driving maneuver and fails to engage in the surveillance type action.

Abstract: Systems and methods are provided for training and suggestion frameworks to improve driver operation of a vehicle. A driver may perform certain maneuvers frequently. Pairing these maneuvers with assigned stimuli may form an association in the driver's mind between the maneuver and the stimuli. A driving scenario may arise in which it would be safer, compliant, more efficient, or more effective for a driver to perform a certain maneuver. If a driver has formed a strong mental association between a maneuver and assigned stimuli, the assigned stimuli may be presented to prompt the driver to perform the maneuver. A driver may still retain ultimate control of a vehicle and may choose not to perform the maneuver.

Abstract: Systems and methods are provided for predicting a vehicle’s motion, wherein a first set of predictions of the vehicle’s motion based on a primary model and a second set of predictions of the vehicle’s motion based on a secondary model are made, an error between a prediction in the first set of predictions and the corresponding prediction in the second set of predictions is determined, a disturbance estimation using the error is generated, and one or more other predictions in the first set of predictions based on the primary model are corrected using the disturbance estimation.

Abstract: A method for prioritizing occlusion information is presented. The method includes determining, at a first time period, a first sensor's view of a spatial area is occluded. The method also includes observing, at a second time period, the spatial area. The method further includes determining a level of risk associated with the spatial area based on the observation. The method still further includes prioritizing transmission of the occlusion information corresponding to the spatial area based on the determined level of risk and transmitting the occlusion information corresponding to the spatial area based on the priority.

Abstract: Described are systems and methods for self-learned label refinement of a training set. In on example, a system includes a processor and a memory having a training set generation module that causes the processor to train a model using an image as an input to the model and 2D bounding based on 3D bounding boxes as ground truths, select a first subset from predicted 2D bounding boxes previously outputted by the model, retrain the model using the image as the input and the first subset as ground truths, select a second set of predicted 2D bounding boxes previously outputted by the model, and generate the training set by selecting the 3D bounding boxes from a master set of 3D bounding boxes that have corresponding 2D bounding boxes that form the second subset.