Abstract: A method is disclosed. The method includes moving a vehicle having a plurality of wheels along a track, receiving a marker data or a marker signal from a marker at a location of the track using a reader of the vehicle, controlling a first rotational speed of a first wheel of the plurality of wheels at the location of the track based on the marker data or the marker signal, and controlling a second rotational speed of a second wheel of the plurality of wheels at the location of the track, independently of the first rotational speed, based on the marker data or the marker signal. The first rotational speed is different from the second rotational speed.

Abstract: An artificial intelligence system is enabled for an autonomous vehicle to calculate a plurality of trajectory paths. The system is enabled to identify and recognize objects. The autonomous vehicle is enabled using a convolutional neural network to determine various paths associated with the autonomous vehicle and also of other moveable objects in the environment. The autonomous vehicle is further enabled to use a cloud based server or edge computing device to pre calculate environments including predictions of environments. Upon an threshold of paths available an aggressive driving move may be authorized to proceed by the autonomous vehicle. The system is enabled to identify and recognize objects using a plurality of sensors including vision-based sensors, image sensors, video sensors, camera sensors, LIDAR, and sensor fusion. Data from various sensors may be combined and fused together to provide enhanced input to the autonomous vehicle.

Abstract: A system and method for detecting tracking vehicles are provided. The method includes determining that a second vehicle is a tracking vehicle for a first vehicle by applying at least one tracking vehicle determination rule to data captured by the first vehicle with respect to the second vehicle, wherein each tracking vehicle determination rule defines a combination of parameters such that the second vehicle is determined to be the tracking vehicle when the data captured by the first vehicle includes the combination of parameters for at least one of the at least one tracking vehicle determination rule; generating a notification upon determination that the second vehicle is the tracking vehicle; and executing a resolution process based on determination that the second vehicle is the tracking vehicle.

Abstract: A vehicle travel assistance apparatus includes a traveling environment information acquiring unit, a preceding vehicle recognizing unit, a vehicle shadow detector, a front vehicle shadow determining unit, and a second preceding vehicle estimating unit. The vehicle shadow detector detects a vehicle shadow of a train of vehicles including a first vehicle shadow of a first preceding vehicle recognized by the preceding vehicle recognizing unit and a front vehicle shadow of a front vehicle traveling in front of the first preceding vehicle based on a brightness difference from a road surface. When the front vehicle shadow determining unit determines that the front vehicle shadow separating from the first vehicle shadow is identified from the vehicle shadow, the second preceding vehicle estimating unit estimates a position of a second preceding vehicle belonging to the train of the vehicles including the first preceding vehicle based on the front vehicle shadow.

Abstract: The method of managing unmanned aerial vehicle (UAV) patrols is a method for organizing and managing a fleet of UAVs for patrolling a geographic region for the detection of threats. During patrols, the UAVs visit two types of waypoints: critical waypoints and strategic waypoints. To establish the strategic waypoints, a set of seed points are generated using a beta probability distribution. Voronoi tessellation is applied to produce the set of strategic waypoints within the region of interest by using the set of seed points as an input. Each of the strategic waypoints has a set of coordinates on the three-dimensional occupancy map associated therewith. The set of coordinates associated with each of the strategic waypoints is a centroid of a corresponding Voronoi cell produced by the Voronoi tessellation. A priority score is assigned to each of the critical waypoints and each of the strategic waypoints.

Abstract: This application discloses a method, an apparatus, and a storage medium for constructing a simulated vehicle lane change trajectory and control a vehicle based on it. One method includes: determining a target preceding vehicle according to a position of a lane change vehicle; determining target longitudinal traveling data; determining at least one lane change trajectory in a determination of at least one of following conditions: whether lane change condition is met; whether it is determined to continue to change lanes; and controlling the lance change vehicle based on the at least one lane change trajectory.

Abstract: A dangerous scene prediction device 80 for predicting a dangerous scene occurring during driving of a vehicle includes a learning model selection/synthesis unit 81 and a dangerous scene prediction unit 82. The learning model selection/synthesis unit 81 selects, from two or more learning models, a learning model used for predicting the dangerous scene, depending on a scene determined based on information obtained during the driving of the vehicle. The dangerous scene prediction unit 82 predicts the dangerous scene occurring during the driving of the vehicle, using the selected learning model.

Abstract: The use of multiple horizon optimization for vehicle dynamics and powertrain control of a vehicle is provided. Long horizon optimization for a trip of the vehicle is performed, and an optimal value function is determined. Data is received from powertrain and/or connectivity features from one or more of components of the vehicle. Short horizon optimization for the trip is performed using a rollout algorithm, the optimal value function, and the received data. The operation of the vehicle is adjusted using results of the short horizon optimization.

Abstract: Techniques are described for determining to modify a coordinate system in response to the occurrence of a condition. As non-limiting examples, such conditions comprise distance traveled, speed of the vehicle (or system), an amount of computational resources available, a time since the last change, a number of objects present, the presence (or absence) of a particular object proximate the vehicle, a distance to a proximate object, based on a particular frequency, or the like. Such modifications may improve the operation and safety of a computing system used for path planning and trajectory generation by allowing lower precision numerical representations to be used in safety-critical situations while avoiding potential computational errors that would otherwise result from using such a representation.

Abstract: In various examples, a 3D surface structure such as the 3D surface structure of a road (3D road surface) may be observed and estimated to generate a 3D point cloud or other representation of the 3D surface structure. Since the representation may be sparse, one or more densification techniques may be applied to densify the representation of the 3D surface structure. For example, the relationship between sparse and dense projection images (e.g., 2D height maps) may be modeled with a Markov random field, and Maximum a Posterior (MAP) inference may be performed using a corresponding joint probability distribution to estimate the most likely dense values given the sparse values. The resulting dense representation of the 3D surface structure may be provided to an autonomous vehicle drive stack to enable safe and comfortable planning and control of the autonomous vehicle.

Abstract: A system for lateral control in-path tracking of an autonomous vehicle includes a lateral controller. The lateral controller controls movement of the autonomous vehicle relative to a path and receives as an input a desired target. An outer control loop of the lateral controller includes a first controller generating an output based on the difference between the desired target and a current position of the autonomous vehicle. An inner control loop of the lateral controller includes a second controller receiving the generated output from the first controller. The inner control loop generates a sideslip angle and a yaw rate, wherein the sideslip angle and the yaw rate are returned to the second controller. The sideslip angle and the yaw rate are used to generate the relative yaw angle and lateral distance, which are returned to the first controller as the current position of the autonomous vehicle.

Abstract: Disclosed is a decision-making and planning integrated method for a nonconservative intelligent vehicle in a complex heterogeneous environment, including the following steps: offline establishing and training a social interaction knowledge learning model; obtaining state data of the traffic participants and state data of an intelligent vehicle online in real time, and splicing the state data to obtain an environmental state; using the environmental state as an input to the trained social interaction knowledge learning model to obtain predicted trajectories of all traffic participants including the nonconservative intelligent vehicle; updating the environmental state based on the predicted trajectories; and inputting the updated environmental state to the social interaction knowledge learning model to complete trajectory decision-making and planning for the nonconservative intelligent vehicle by iteration, where a planned trajectory of the nonconservative intelligent vehicle is a splicing result of a first poi

Abstract: An unmanned vehicle path control method, apparatus, and system queries the current position and a target position of an unmanned vehicle. Path planning according to the current position, the target position, and a heading direction indicated by landmarks arranged on two sides of a road is performed, so as to obtain a planned path having the minimum path cost. The unmanned vehicle travels forward on a predetermined side of the road and sends the planned path to the unmanned vehicle, so that the unmanned vehicle reaches the target position according to the planned path.

Abstract: Disclosed is a nighttime cooperative positioning method based on an unmanned aerial vehicle (UAV) group, falling within the technical field of aircraft navigation and positioning. According to the present disclosure, the cooperative visual positioning and the collision warning for UAVs are realized by means of light colors of the UAVs, respective two-dimensional turntable cameras and a communication topology network, without adding additional equipment and without relying on an external signal source, avoiding external interference. Compared with the positioning method in a conventional manner, in the present disclosure, the system is effectively simplified, and the cooperative positioning among the interiors of a UAV cluster can be realized relatively simply and at a low cost to maintain the formation of the UAV group.

Abstract: The present disclosure is directed to generating vehicle motion corridors for use in generating autonomous vehicle paths. In particular, a computing system can access map data for a geographic area and sensor data for the geographic area around an autonomous vehicle. The computing system can identify, based on the map data and the sensor data, object data describing a position and a size of one or more objects in the geographic area of the autonomous vehicle. The computing system can access path data describing a nominal path through the geographic area. The computing system can determine a plurality of corridor segments associated with the nominal path. The computing system can generate a vehicle motion corridor by aggregating the plurality of corridor segments, wherein the vehicle motion corridor defines an area in which the autonomous vehicle can travel without colliding with the one or more objects.

Abstract: Among other things, techniques are described for identifying sensor data from a sensor of a first vehicle that includes information related to a pose of at least two other vehicles on a road. The technique further includes determining a geometry of a portion of the road based at least in part on the information about the pose of the at least two other vehicles. The technique further includes comparing the geometry of the portion of the road with map data to identify a match between the portion of the road and a portion of the map data. The technique further includes determining a pose of the first vehicle relative to the map data based at least in part on the match.

Abstract: Industrial machines at a mine site or other worksite are monitored and controlled using a multi-phase implementation, based on data received from multiple data sources at different times. Machine telemetry data or other sensor data received from a first machine, in combination with additional sensor data from other machines at the site, detailed material attribute data received from external data sources, and/or outputs from trained machine-learned models, are used to determine characteristics of a material load and/or material blend, and to control machines at the mine site. A control system performs a multi-phase implementation for monitoring machines and/or load data in response to data received at different times, updating the material load and blend characteristics, and controlling machines at the site based on the characteristics of the material load or material blend.

Abstract: Systems and methods for energy efficient mobility are provided. The method may include identifying a vehicle moving on a road, calculating a first energy required to complete a trip, determining a size of an air pocket zone of the vehicle, calculating a second energy required to transition to the air pocket zone of the vehicle and to complete the trip in the air pocket zone of the vehicle, and transitioning to the air pocket zone of the vehicle if the second energy is more energy efficient than the first energy.

Abstract: The invention discloses a method and a system for state feedback predictive control of a high-speed train based on a forecast error. The method comprises: obtaining a speed prediction model of a high-speed train y ^ k + p = C ? A p ? x k + ? i = 1 p ? C ? A i - 1 ? B ? u k + p - i ; predicting speeds of the train at times k and k+p according to the speed prediction model; obtaining an actual speed output value of the train; determining a speed prediction error at time k according to the prediction speed of the train and the actual speed output value of the train; correcting a prediction speed of the train at time k+p, according to the speed prediction error, to obtain a corrected prediction speed of the train; calculating a control force uk of the train according to uk=??1(p)[yk+pr?yk?Kxk+?k]; and applying a control force to the train based on the control force uk.

Abstract: The present disclosure enables a collection of training data suitable for training an autonomous driving system of a vehicle. A database is provided that stores predefined road scenarios, and user devices are provided with a simulation game for controlling a vehicle agent in a road scenario. A plurality of user devices, running the simulation game, play a road scenario from the database to control a vehicle agent in the simulation game with steering actions entered by individual users, which generates a human demonstration of each road scenario that is played. Several human demonstrations played on the plurality of user devices can be maintained by a demonstration database and made available as training data suitable for training an autonomous driving system of a vehicle. A large amount of training data can be automatically generated, without requiring expensive or time-consuming real-world tests to generate suitable training data manually.