Abstract: A pedestrian tracking system includes: a buffer or a memory configured to store a trajectory sequence of a pedestrian; a step attention module and a control module. The step attention module iteratively performs a step attention process to predict states of the pedestrian. Each iteration of the step attention process includes the step attention module: learning the stored trajectory sequence to provide time-dependent hidden states, reshaping each of the time-dependent hidden states to provide two-dimensional tensors; condensing the two-dimensional tensors via convolutional networks to provide convolutional sequences; capturing global information of the convolutional sequences to output a set of trajectory patterns represented by a new sequence of tensors; learning time-related patterns in the new sequence and decoding the new sequence to provide one or more of the states of the pedestrian; and modifying the stored trajectory sequence to include the predicted one or more of the states of the pedestrian.

Abstract: The present disclosure is directed toward a method that includes acquiring current dynamic characteristics of a moving object about an un-signalized intersection, determining a dynamic traffic flow of the un-signalized intersection based on the current dynamic characteristics to determine a position of the moving object about the un-signalized intersection, and predicting future dynamic characteristics of the moving object based on the dynamic traffic flow and a predictive control. The future dynamic characteristics includes a predicted position of the moving object at a predefined time in the future. The method further calculates an entry parameter for a vehicle about an entrance of the un-signalized intersection based on the dynamic traffic flow and the predicted future dynamic characteristics and notifies the vehicle of the entry parameter.

Abstract: A RLP system for a host vehicle includes a memory and levels. The memory stores a RLP algorithm, which is a multi-agent collaborative DQN with PER algorithm. A first level includes a data processing module that provides sensor data, object location data, and state information of the host vehicle and other vehicles. A second level includes a coordinate location module that, based on the sensor data, the object location data, the state information, and a refined policy provided by the third level, generates an updated policy and a set of future coordinate locations implemented via the first level. A third level includes evaluation and target neural networks and a processor that executes instructions of the RLP algorithm for collaborative action planning between the host and other vehicles based on outputs of the evaluation and target networks and to generate the refined policy based on reward values associated with events.

Abstract: The present disclosure is directed toward a method that includes acquiring current dynamic characteristics of a moving object about an un-signalized intersection, determining a dynamic traffic flow of the un-signalized intersection based on the current dynamic characteristics to determine a position of the moving object about the un-signalized intersection, and predicting future dynamic characteristics of the moving object based on the dynamic traffic flow and a predictive control. The future dynamic characteristics includes a predicted position of the moving object at a predefined time in the future. The method further calculates an entry parameter for a vehicle about an entrance of the un-signalized intersection based on the dynamic traffic flow and the predicted future dynamic characteristics and notifies the vehicle of the entry parameter.

Abstract: The present disclosure is directed toward a method that includes defining a roundabout path plan for a subject vehicle based on roundabout characteristics and dynamic characteristics, among other data. The method further determines whether a platoon can be formed with at least one surrounding vehicle based on the roundabout path plan, and calculates an entry parameter for the platoon based on dynamic traffic flow of the roundabout and a predictive control in response to determining that the platoon can be formed. The method collaborates and verifies the entry parameter with the platoon to obtain an agreed entry parameter.

Abstract: A sensor data anomaly detection system for a vehicle includes one or more processors and a memory in communication with the one or more processors that stores an anomaly detection module and an anomaly correction module. The anomaly detection module causes the one or more processors to analyze one or more signals from at least one sensor of the vehicle for at least one potential anomaly and compare correlating information from one or more external sources to the at least one potential anomaly to confirm when the potential anomaly is an actual anomaly The anomaly correction module causes the one or more processors to correct the actual anomaly in the one or more signals to generate a corrected signal when the actual anomaly is present.

Abstract: A RLP system for a host vehicle includes a memory and levels. The memory stores a RLP algorithm, which is a multi-agent collaborative DQN with PER algorithm. A first level includes a data processing module that provides sensor data, object location data, and state information of the host vehicle and other vehicles. A second level includes a coordinate location module that, based on the sensor data, the object location data, the state information, and a refined policy provided by the third level, generates an updated policy and a set of future coordinate locations implemented via the first level. A third level includes evaluation and target neural networks and a processor that executes instructions of the RLP algorithm for collaborative action planning between the host and other vehicles based on outputs of the evaluation and target networks and to generate the refined policy based on reward values associated with events.

Abstract: Systems and methods are disclosed and include a node storing a local table and a control module that collects heterogeneous sensor data, tracks objects in an environment of the node, and converts the heterogeneous sensor data into time series data. An object-relational mapping module formats the time series data for direct compatibility to data in a master database of a cloud-based server. A correlation module further formats the time series data to provide resultant data in a single format commonly used by multiple map sources. A first primitive extraction module extracts primitives from the resultant data. An edge computing module generates a first table of data corresponding to the primitives and compatible with a master table of the master database, detects a changed primitive, uploads the changed primitive to the cloud-based server, receives updated data based on the changed primitive, and updates the local table based on the updated data.

Abstract: The present disclosure provides a dimension estimating system for a host vehicle. The dimension estimating system includes an image sensor and a dimension estimator. The image sensor obtains image data of at least one side of a target vehicle around the host vehicle. The dimension estimator estimates, through a machine learning algorithm, a dimension of the target vehicle based on the image data of the one side of the target vehicle obtained by the image sensor.

Abstract: An AI system is provided and includes a long short term memory and data processing, feature selection, behavior recognition, parameter, and driver assistance modules. The data processing module, at a host vehicle: receives training data associated with features; based on the training data, determines candidate features using a synchronous sliding window; and indexes and ranks the candidate features. The feature selection module selects ones of the candidate features having higher priority than other ones of the candidate features and filters the training data of the selected features to provide filtered data. The behavior recognition module determines a behavior of a remote vehicle based on the filtered data. The parameter module determines a parameter of the remote vehicle based on the determined behavior. The long short term memory predicts a value of the parameter. The driver assistance module assists in operation of the host vehicle based on the predicted value.

Abstract: The present disclosure provides an auto-braking system which includes a braking device, a receiver, and a controller. The controller includes a determining portion, a verifying portion and an executing portion. The determining portion determines whether a vehicle will violate a red signal of the traffic light based on the traffic light state information. The verifying portion verifies the traffic light state information is valid by comparing the traffic light state information with the image of the traffic light captured by the sensor. The executing portion executes a first braking control to slow the vehicle at a first deceleration rate when the determining portion determines that the vehicle will violate a red signal. The executing portion executes a second braking control to slow the vehicle at a second deceleration rate to stop at a specified position when the verifying portion verifies that the traffic light state information is valid.

Abstract: The present disclosure provides an auto-braking system which includes a braking device, a receiver, and a controller. The controller includes a determining portion, a verifying portion and an executing portion. The determining portion determines whether a vehicle will violate a red signal of the traffic light based on the traffic light state information. The verifying portion verifies the traffic light state information is valid by comparing the traffic light state information with the image of the traffic light captured by the sensor. The executing portion executes a first braking control to slow the vehicle at a first deceleration rate when the determining portion determines that the vehicle will violate a red signal. The executing portion executes a second braking control to slow the vehicle at a second deceleration rate to stop at a specified position when the verifying portion verifies that the traffic light state information is valid.

Abstract: A silicon carbide semiconductor device such as JFET, SIT and the like is provided for accomplishing a reduction in on-resistance and high-speed switching operations. In the JFET or SIT which turns on/off a current with a depletion layer extending in a channel between a gate region formed along trench grooves, a gate contact layer and a gate electrode, which can be supplied with voltages from the outside, are formed on one surface of a semiconductor substrate or on the bottom of the trench groove. A metal conductor (virtual gate electrode) is formed in ohmic contact with a p++ contact layer of the gate region on the bottom of the trench grooves independently of the gate electrode. The virtual gate electrode is electrically isolated from the gate electrode and an external wire.

Abstract: A method of forming an ohmic contact on a substrate composed of a wide-band gap semiconductor material includes: depositing a transition metal group metal on the substrate; annealing the substrate at a high temperature to cause a solid state chemical reaction between the substrate and the deposited metal that forms a modified layer in the substrate having modified properties different than the substrate, and by-products composed of a silicide and a nanocrystalline graphite layer; selectively etching the substrate to remove one or more of the by-products of the solid state chemical reaction from a surface of the substrate; and depositing a metal film composed of a transition group metal over the modified layer on the substrate to form the ohmic contact. The modified layer permits formation of the ohmic contact without high temperature annealing after depositing the metal film.

Abstract: A semiconductor device includes (a) a vertical field effect transistor, the vertical field effect transistor including a drain electrode formed on a first surface of a first conductivity type of a semiconductor, a pair of first trenches formed from a second surface of the semiconductor, control regions of a second conductivity type formed respectively along the first trenches, a source region of the first conductivity type formed along the second surface of the semiconductor between the first trenches, a source electrode joined to the source region, and a gate electrode adjacent to the control regions, (b) a pair of second trenches formed from the second surface of the semiconductor independently of the field effect transistor, (c) control regions of the second conductivity type formed along the second trenches, and (d) a diode having a junction formed on the second surface between the second trenches.

Abstract: A silicon carbide semiconductor device such as JFET, SIT and the like is provided for accomplishing a reduction in on-resistance and high-speed switching operations. In the JFET or SIT which turns on/off a current with a depletion layer extending in a channel between a gate region formed along trench grooves, a gate contact layer and a gate electrode, which can be supplied with voltages from the outside, are formed on one surface of a semiconductor substrate or on the bottom of the trench groove. A metal conductor (virtual gate electrode) is formed in ohmic contact with a p++ contact layer of the gate region on the bottom of the trench grooves independently of the gate electrode. The virtual gate electrode is electrically isolated from the gate electrode and an external wire.

Abstract: A method for manufacturing a silicon carbide semiconductor device includes the steps of: forming a trench mask on an upper surface of a semiconductor substrate; forming the trench such that the trench having an aspect ratio equal to or larger than 2 and having a trench slanting angle equal to or larger than 80 degrees is formed; and removing a damage portion in such a manner that the damage portion disposed on an inner surface of the trench formed in the semiconductor substrate in the step of forming the trench is etched and removed in hydrogen atmosphere under decompression pressure at a temperature equal to or higher than 1600° C.