Abstract: A method for determining scene risk for an autonomous vehicle includes receiving scene data, predicting risk scores using a trained risk prediction model, determining an appropriate autonomous vehicle behavior based on these risk scores, and controlling the vehicle accordingly. In some implementations, the method includes training the risk prediction model by generating initial risk scores from a labeled training dataset, learning an intermediate model from these scores, and training the risk prediction model using multimodal training data and secondary risk scores.

Abstract: A posture determination method is to determine, based on an image acquired for a component having a plurality of surfaces, a posture of the component. This posture determination method includes recognizing an outer shape of the component from the image, detecting a feature amount of a surface shape in a determination area being a part of the component based on the outer shape of the component recognized in the recognizing, and determining. In the determining, the posture of the component is determined by comparing a first reference amount indicating a reference feature amount in the determination area of a first surface of the component and a second reference amount indicating a reference feature amount in the determination area of a second surface of the component with the feature amount detected in the detecting.

Abstract: A grounds maintenance system comprising: a robot tractor comprising; a robot body; a drive system including one or more motorized drive wheels to propel the robot body; a control system coupled to the drive system, the control system configurable to store a mow plan that specifies a set of paths to be traversed for a grounds maintenance operation and control the drive system to autonomously traverse the set of paths to implement the mow plan; a battery system comprising one or more batteries housed in the robot body; and a low-profile mowing deck coupled to the robot body, the mowing deck adapted to tilt and lift relative to the robot body, wherein the control system is configured to control tilting and lifting of the mowing deck and cutting by the mowing deck.

Abstract: A system and method for operation of an autonomous vehicle (AV) yard truck is provided. A processor facilitates autonomous movement of the AV yard truck, and connection to and disconnection from trailers. A plurality of sensors are interconnected with the processor that sense terrain/objects and assist in automatically connecting/disconnecting trailers. A server, interconnected, wirelessly with the processor, that tracks movement of the truck around and determines locations for trailer connection and disconnection. A door station unlatches/opens rear doors of the trailer when adjacent thereto, securing them in an opened position via clamps, etc. The system computes a height of the trailer, and/or if landing gear of the trailer is on the ground and interoperates with the fifth wheel to change height, and whether docking is safe, allowing a user to take manual control, and optimum charge time(s). Reversing sensors/safety, automated chocking, and intermodal container organization are also provided.

Abstract: Disclosed herein is an autonomous vehicle comprising at least one subsystem capable of determining a location corresponding to a connectable tool; determining an orientation for interconnecting with the connectable tool; aligning the autonomous vehicle for interconnecting with the connectable tool based on the location and the orientation of the connectable tool; engaging the connectable tool at one or more physical connection points and at least one electrical connection point; and establishing power line modem (PLM) communications between the autonomous vehicle and the connectable tool via an electrical power connection resulting from the engaging the at least one electrical connection point; and, before commencing full power operations over said electrical power connection, determining that the electrical power connection is without current diversion by transmitting a fixed amount of non-lethal electrical power from the autonomous vehicle to the connectable tool via the electrical power connection and dete

Abstract: A controller may be configured to receive one or more measured rotational speeds of a wheel of a vehicle. The controller may be configured to determine whether the one or more measured rotational speeds are unreliable relative to one or more previous rotational speeds of the wheel of the vehicle. The controller may be configured to calculate a replacement rotational speed of the wheel and use the replacement rotational speed of the wheel to control or restrict movement of the vehicle using or based on the replacement rotational speed in place of the one or more measured rotational speeds.

Abstract: An effector system comprises: an effector assembly including: a first segment having a conduit between a proximal end and a distal end, an inner surface of the conduit having a conductive portion and a resistive portion; an output member slidable within the conduit, the output member having a conductive exterior contacting the inner surface; a first shape-memory transducer affixed between the proximal end and the output member; and a second shape-memory transducer affixed between the distal end and the output member; and a controller configured to: selectively energize: (i) the first shape-memory transducer to slide the output member in a first direction, (ii) the second shape-memory transducer to slide the output member in a second direction, or (iii) the resistive portion of the inner surface; and in response to energizing the resistive portion, determine a position of the output member.

Abstract: Provided are systems and techniques for a medical procedure. For example, the system may include one or more robotic arms, an imaging device, a master controller, a viewer configured to render one or more digital images based on image data from the imaging device, at least one computer-readable memory having stored thereon executable instructions, and one or more processors. The one or more processors may be configured to execute the instructions to cause the system to: in a first mode of operation, cause movement of at least one of the robotic arms; and in a second mode of operation, cause the viewer to display an interactive menu and a graphical overlay on the one or more digital images.

Abstract: A vehicle motion manager includes one or more processors configured to: receive a first acceleration request value that requires degeneration and a second acceleration request value that does not require the degeneration from a first application and a second application, respectively; select a smallest acceleration request value out of the acceleration request values received by the one or more processors; when the one or more processors select the first acceleration request value, repeatedly output a degeneration value based on the first acceleration request value; and output an instruction value corresponding to the acceleration request value selected by the one or more processors or the degeneration value to an actuator of a vehicle. The one or more processors are configured to, in response to selection of the first acceleration request value, receive the degeneration value output by the one or more processors instead of the first acceleration request value.

Abstract: Disclosed is an acceleration limit function relaxation apparatus. The apparatus may determine, based on information indicating manipulation of a brake pedal of a vehicle during an acceleration limit mode, that the vehicle is in a temporary braking state, determine a vehicle speed, measured at a start of the manipulation of the brake pedal, as a target vehicle speed for reacceleration, determine, based on information indicating manipulation of an accelerator pedal, that the vehicle is reaccelerated, determine an acceleration limit relaxation factor based on a cumulative braking amount calculated by measuring amounts of braking during the temporary braking state, and cause acceleration limit relaxation of the vehicle by applying the acceleration limit relaxation factor to an acceleration limit of the vehicle.

Abstract: According to the present invention, provided is a robot control device that can improve relatively easily the positioning accuracy of a robot.

Abstract: Aspects of the disclosure provide controlling an autonomous vehicle. For example, a first trajectory generated by a first planning system may be received. A portion of the first trajectory may be input into a second planning system. A second trajectory generated by the second planning system based on the portion of the first trajectory may be received. The second trajectory may be used to determine whether to validate the first trajectory, and when the second trajectory is determined to be validated, the autonomous vehicle may be enabled to be controlled based on the first trajectory.

Abstract: Systems/techniques that facilitate artificially intelligent provision of post-vehicular-collision evidence are provided. In various embodiments, a system can be onboard a vehicle and can capture, via one or more cameras or microphones of the vehicle, vicinity data associated with a vicinity of the vehicle. In various instances, the system can generate, via execution of a deep learning neural network on the vicinity data, a classification label indicating whether a vehicular collision not involving the vehicle has occurred in the vicinity of the vehicle. In various cases, the system can record, in response to the classification label indicating that the vehicular collision has occurred and via the one or more cameras or the one or more microphones, post-collision evidence associated with the vicinity of the vehicle. In various aspects, the system can broadcast the classification label and the post-collision evidence to an emergency service computing device.

Abstract: A processor unit (3) for model-based predictive control of a vehicle (1) taking into account an arrival time factor is configured to calculate a trajectory for the vehicle (1) based at least in part on at least one arrival time factor, with the trajectory including an entire route (20) to a specified destination (19) at which the vehicle (1) is to arrive, and with the at least one arrival time factor influencing an arrival time of the vehicle (1) at the specified destination (19). Additionally, the processor unit (3) is configured to optimize a section of the trajectory for the vehicle (1) for a sliding prediction horizon by executing a model-based predictive control (MPC) algorithm (13), where the MPC algorithm (13) includes a longitudinal dynamic model (14) of a drive train (7) of the vehicle (1) and a cost function (15) to be minimized.

Abstract: A method for inducing a vehicle occupant to bring a vehicle to a stop includes receiving malfunction data. The malfunction data is indicative of an impending vehicular malfunction will occur in a vehicle; determining that the impending vehicular malfunction will occur using the malfunction data. The method also includes determining that the vehicle is capable of operating without hazards for a predetermined amount of time in response to determining that the impending vehicular malfunction will occur. Further, the method includes commanding an actuator of the vehicle to provide a sensory feedback to the vehicle occupant to induce the vehicle occupant to stop the vehicle in response to determining that the impending vehicular malfunction will occur and determining that the vehicle is capable of operating without hazards for a predetermined amount of time.

Abstract: A robot is provided. The robot includes a sensor, a driver, a communication interface, and a processor configured to acquire first context data through the sensor, receive second context data acquired by at least one other robot through the communication interface, identify at least one context data of the first context data and the second context data based on a collaboration scenario of the robot and the other robot, input the identified at least one context data to a predetermined task allocation algorithm related to the collaboration scenario to acquire task information corresponding to the robot, and control the driver based on the acquired task information.

Abstract: Disclosed herein is autonomous vehicle charging station featuring self-cleaning contact points, power flow safety features, and electrical connection testing (meeting minimum performance thresholds) for safe and efficient power transfer from the charging station to the autonomous vehicle. After docking, power line modem (PLM) communications are established between the autonomous vehicle and the charging station via an electrical power connection resulting from the docking act as an initial test of connectivity, and then the electrical power connection is determined to be safe with non-lethal power flow before charging the autonomous vehicle.

Abstract: Disclosed are solutions for an autonomous vehicle to perform an initial navigational setup at an operating site by: receiving initial location and orientation data pertaining to the autonomous vehicle; iteratively sensing a plurality of recognizable objects and determining a plurality of fixed reference locations corresponding to each of the plurality of recognizable objects relative to the initial location and orientation data; and developing an operational path plan for the autonomous vehicle to traverse the operating site based on the plurality of fixed reference locations. The autonomous vehicle may be an autonomous mower, the dynamic object may be a solar panel and/or solar panel post, and the sensing may be performed using light detection and ranging (LIDAR).

Abstract: Disclosed are solutions for proactively sensing the condition of the ground in the path of the autonomous vehicle in order to mitigate the risks of an autonomous vehicle encountering troublesome terrain, navigation obstacles, and other potential risks to the continued uninterrupted operation of said autonomous vehicle. Accordingly, various implementations disclosed herein are directed to the use of ground analysis sensors to assist in the detection of navigationally difficult terrain including, for example, the use of cantilevered sensors operating well in front of the autonomous vehicle's front wheels (drive wheels or otherwise).

Abstract: Disclosed are solutions for an autonomous vehicle to real-time detect and determine dynamic and/or obscured obstacles to support navigation and services to within a desired proximity of said obstacles. Certain such implementations are specifically directed to autonomous mowers, for example, capable of real-time object detection to determine location and orientation of solar panels in a solar farm, for example, and based on the location and orientation of such solar panels further determine the location of their corresponding posts that may be otherwise obstructed from direct detection by the autonomous mower's other sensing systems possibly due to vegetation growth around the posts or other reasons.