Denso International America, Inc.

Denso Corporation

Haris Volos

Takashi Bando

Yunfei Xu

Ravi Akella

Kenji Konishi

Kazuhito Takenaka

Hideaki Misawa

Yuki Shinohara

Takafumi Okayasu

An example computer system includes memory hardware configured to store a machine learning model and historical driving data vector inputs, and processor hardware configured to execute instructions including training the machine learning model with the historical driving data vector inputs to generate a reconstructed driving data output, wherein the reconstructed driving data output includes at least one reconstruction error score indicative of a likelihood that a driving data input corresponds to a desired driving behavior, obtaining a current driving data input, and supplying the current driving data input to the machine learning model to generate a reconstruction error score based on the current driving data input. The instructions may include determining a driving score according to the reconstruction error score, identifying at least one driver notification according to the determined driving score, and transmitting the identified at least one driver notification to a computing device or display.

Systems and machine learning models for automated vehicle monitoring

A system for setting parameters within a first vehicle for a first user of the first vehicle is provided and includes a memory, transceiver, and a processing module. The memory stores normalized parameters, which are specific to the first user. Each of the normalized parameters is a parameter of the vehicle. The transceiver receives the normalized parameters from a mobile device of the first user, a server in a cloud-based network, or a portable memory device. The processing module: determines whether the first user is authorized for parameter translation services; if the first user is authorized for parameter translation services, translates the normalized parameters to resultant parameters; generates parameter recommendations based on the resultant parameters; presents the parameter recommendations to the first user; and adjusts current parameters of the first vehicle based on the resultant parameters and a received response from the first user with regards to the parameter recommendations.

Vehicle recommendation and translation system for setting personalized parameters within vehicles of mobility sharing environments

A central latency system includes a communication device configured to exchange data with an external device. The data includes current latency data, current contextual data associated with the latency data, or a combination thereof. The system further includes a controller configured to aggregate latency data in response to acquiring the current latency data, generate a latency characterization information based on the aggregated latency data, and transmit the latency characterization information to the external device.

System and method for collaborative centralized latency characterization

A vehicle application enabling system is provided and includes a latency characterization and prediction module and an application enable module. The latency characterization and prediction module includes: an estimate sample module configured to collect and store obtained latency estimates, where the obtained latency estimates are associated with transmission of signals in a network for one or more vehicle applications; a sample characterization module configured to characterized the obtained latency estimates; a trend module configured to determine a trend based on the characterized obtained latency estimates; a point projection module configured to generate projected latency estimates based on the trend; and a projection characterization module configured to characterize the projected latency estimates. The application enable module is configured to determine whether to enable the one or more vehicle applications based on the characterization of the projected latency estimates.

Vehicle application enabling and network routing systems implemented based on latency characterization and projection

A vehicle application enabling system is provided and includes a memory and initialization, latency evaluation, and application enable modules. The initialization module: receives a maximum network latency; sets a percentage of occurrences that the maximum network latency is not satisfied, a maximum false positive rate, and a maximum deviation value; and calculates a weighting factor based on the percentage of occurrences, maximum false positive rate and maximum deviation value. The latency evaluation module implements a latency evaluation algorithm, which includes: comparing one or more latency estimates to the maximum network latency to provide one or more samples; updating confusion matrix statistics based on the one or more samples; updating a probability threshold based on the maximum false positive rate; updating weighted observations based on the weighting factor; and determining a predicted decision based on the probability threshold. The application enable module executes the vehicle application based on the probability threshold.

Application specific connected service provision system

A travel route selection system is provided and includes historical latency data and travel route modules. The historical latency data module includes data and characterization modules. The data module collects historical latency data. The historical latency data are associated with transmission of signals in a network for one or more vehicle applications of a vehicle. The characterization module: obtains characteristic distributions of the historical latency data at selected locations along travel routes; based on the characteristic distributions, (i) calculates metrics along the travel routes for the vehicle, or (ii) combines the characteristic distributions to obtain an overall distribution for each of the travel routes; and based on the metrics or the overall distribution, performs a statistical process to predict latencies of the signals along each of the travel routes. The travel route module selects one of the travel routes based on the predicted latencies of the signals.

Travel routing selection system and methods implemented based on characterization of historical latency data

An automatic external incident detection and reporting system for a vehicle includes at least one of a plurality of cameras and a plurality of proximity sensors, an incident determination unit, a remote vehicle or object position determination unit, and a communication unit. The incident determination unit is configured to receive signals from the at least one of the cameras and proximity sensors, detect whether an external incident has occurred, and determine a type of incident. The remote vehicle or object position determination unit is configured to receive signals from the at least one of the cameras and proximity sensors and determine an incident location. The communication unit is configured to transmit at least the incident location and the type of incident to remote vehicles, to a local access point, or to share the incident location and the type of incident through a crowd-sourcing manner if the external incident has occurred.

Automatic crowd sensing and reporting system for road incidents

An automatic traffic incident detection and reporting system for a vehicle may include at least one of a plurality of cameras and a plurality of proximity sensors, an incident determination unit, a remote vehicle position determination unit, and a communication unit. The incident determination unit is configured to receive signals from the at least one of the plurality of cameras and the plurality of proximity sensors, detect whether incident involving at least one remote vehicle has occurred, and categorize the incident. The remote vehicle position determination unit is configured to receive signals from the at least one of the plurality of cameras and the plurality of proximity sensors and determine a location of the incident. The communication unit is configured to transmit data related to the incident to at least one of a cloud server and an emergency service provider.

Automatic traffic incident detection and reporting system

A risk assessment system is provided and includes a behavior description generator, driving behavior comparator, and risk assessment modules and a transceiver. The behavior description generator module receives a set of driving features and topics corresponding to a driving behavior of a first driver in a location and under a set of driving conditions. The driving behavior comparator module: compares the set of driving features and the topics to other sets of features and topics, which correspond to driving behaviors of other drivers for the location and same or similar driving conditions as the set of driving conditions; and generates an anomaly score for the first driver based on results of the comparison. The risk assessment module calculates a risk assessment score for the driver based on the anomaly score, where the risk assessment score is indicative of a risk level of the first driver relative to the other drivers.

Growing community of inventors

Sunnyvale, CA, United States of America

Haris Volos