Toyota Motor Engineering & Manufacturing North America, Inc.

Other

Chang-Heng Wang

Onur Altintas

Takamasa Higuchi

Seyhan Ucar

Takayuki Shimizu

Enes Krijestorac

Prashant Tiwari

John P Kenney

Rui Guo

Hongsheng Lu

Xuewei Qi

Ceyhun D Ozkaptan

Dylan J Lebedin

Isfar Tariq

Systems and methods for vehicular-network-assisted federated machine learning are disclosed herein. One embodiment transmits first metadata from a connected vehicle to at least one other connected vehicle; receives, at the connected vehicle, second metadata from the at least one other connected vehicle; receives, at the connected vehicle based on analysis of the first and second metadata, a notification that the connected vehicle has been elected to participate in the current training phase of a federated machine learning process; receives, at the connected vehicle, instructions to prepare the connected vehicle for the next training phase; trains a machine learning model to perform a task at the connected vehicle during the current training phase to produce a locally trained machine learning model; and submits the locally trained machine learning model for aggregation with at least one other locally trained machine learning model to produce an aggregated locally trained machine learning model.

Systems and methods for vehicular-network-assisted federated machine learning

A method includes receiving first environment information related to a first vehicular task from a host vehicle, comparing the first environment information to second environment information captured when a member vehicle performed a second vehicular task corresponding to the first vehicular task using a second set of operating criteria, and determining a first set of operating criteria for performing the first vehicular task based on a similarity score between the first environment information and the second environment information and a success or accuracy rate of the second vehicular task performed by the member vehicle.

Systems and methods for determining operating criteria for performing vehicular tasks

One example embodiment provides one or more of: measuring, via a user equipment (UE), sidelink received signal strength indicator (SL-RSSI) values and sidelink reference signal power received (SL-RSRP) values from a current window of a communication channel shared among a plurality of UEs, measuring, via the UE, the SL-RSRP values from a frequency-domain and a time-domain of a current window of a communication channel shared among the plurality of UEs, excluding a first subset of communication resources from a next window of the communication channel based on the RSSI values in the current window, excluding a first subset of communication resources from a next window of the communication channel based on the SL-RSRP values from a time-domain, excluding a second subset of communication resources from the next window of the communication channel based on the RSRP values, excluding a second subset of communication resources from the next window of the communication channel based on the SL-RSRP values from a frequency-domain, and selecting and reserving resource(s) remaining in the next window of the communication channel after exclusion of the first and second subsets of communication resources from the next window.

Resource selection for 5G NR V2X communications

A system for determining an existence of a change in a region can include a processor, a communications device, and a memory. The memory can store a change determination module and a communications module. The change determination module can cause the processor to: (1) determine, in response to having obtained an indication of a possible change in the region, a location of a connected car in a vicinity of the region and facts about a sensor disposed on the connected car and (2) determine a strategy to obtain data about the possible change. The strategy can be based on at least one of the location of the connected car or the facts about the sensor. The communications module can cause the processor to transmit, via the communications device and to the connected car, an instruction to control, according to the strategy, the sensor to obtain the data.

Determining an existence of a change in a region

System, methods, and embodiments described herein relate to dynamically generating a wide field-of-view three-dimensional pseudo point cloud of an environment around a vehicle. A disclosed method may include capturing, via a camera, a first view in a first image, determining a first depth map based on the first image, obtaining, from an external system, a second image of a second view that overlaps the first view and a second depth map based on the second image, inputting the first image and the second image into a self-supervised homograph network that is trained to output a homographic transformation matrix between the first image and the second image, and generating a three-dimensional pseudo point cloud that combines the first depth map and the second depth map based on the homographic transformation matrix.

Cooperative pseudo point cloud for connected vehicles

A system for exploiting a non-transmission-designated interval in a cycle of a protocol can include a processor and a memory. The memory can store a retrieval module, an alternative transmission determination module, and a schedule module. The retrieval module can cause the processor to retrieve, by a specific node associated with a set of nodes in a network, information about the set. The information can include, for a node in the set, identifications of a frequency and a time slot of a transmission-designated interval in a cycle of the protocol. The alternative transmission determination module can cause the processor to determine, based on the information, an existence, in the set, of a non-transmission-designated interval that has a duration greater than a threshold duration. The schedule module can cause the processor to schedule, in response to the existence, the specific node to produce a transmission during the non-transmission-designated interval.

Exploiting a non-transmission-designated interval in a cycle of a protocol

The disclosure includes embodiments for resolving vehicle application version differences for connected vehicles. A method includes determining a common vehicle application that is installed in both a remote connected vehicle and an ego vehicle and identifying version differences in a common vehicle application installed in both the ego vehicle and the remote connected vehicle. The method includes forming a vehicular micro cloud. The method includes determining a maximum possible functionality of the common vehicle application. The method includes determining a set of tasks to be completed to achieve the maximum possible functionality. The method includes assigning a first subset of the set of tasks to the ego vehicle and a second subset of the set of tasks to the remote connected vehicle. The method includes using the vehicular micro cloud to cause the ego vehicle to complete the first subset and the remote connected vehicle to complete the second subset.

Resolving vehicle application version differences

The disclosure includes embodiments for spinning out members of a first vehicular micro cloud to form a second vehicular micro cloud. According to some embodiments, a method includes forming a first vehicular micro cloud including a set of members. The method includes exiting the first vehicular micro cloud by a subset of the members of the first vehicular micro cloud. The method incudes identifying a reason to form a second vehicular micro cloud by the subset. The method includes executing steps, by a processor and a wireless radio, to form the second vehicular micro cloud by the subset based on the identification of the reason so that, by the formation of the second vehicular micro cloud, the spinning out of the subset of the members of the first vehicular micro cloud is achieved. The processor and the wireless radio are elements of a hub of the second vehicular micro cloud.

Spinning out vehicular micro clouds

The disclosed subject matter relates to managing a vehicular micro cloud. A method may include determining join/leave protocols for a vehicular micro cloud and transmitting the join/leave protocols to a local vehicle in a vicinity of the vehicular micro cloud prior to the local vehicle joining the vehicular micro cloud. The join/leave protocols can define at least: 1) a procedure for the local vehicle to join the vehicular micro cloud and contribute computing resources to a collaborative micro cloud task, and 2) a protocol for handing an incomplete task when the local vehicle leaves the vehicular micro cloud.

Systems and methods to regulate joining and leaving a vehicular micro cloud

Systems, methods, and other embodiments described herein relate to improving the identification of available parking spaces by inferring when a vehicle is likely to depart. In one embodiment, a method includes, responsive to acquiring sensor data about a surrounding environment including at least one parked vehicle, extracting a feature from the sensor data about a context of the parked vehicle. The method includes analyzing the feature to determine a status of the vehicle in relation to whether the vehicle is remaining parked. The method includes providing the status to inform additional vehicles about an availability of a parking spot of the parked vehicle.

Systems and methods for inferring parking departures

A method may include determining whether a plurality of vehicles, each having an application installed are expected to be present within a predetermined distance of a location of an event. The method may further include comparing versions of the applications among the plurality of vehicles and, in response to determining that the plurality of vehicles are expected to be present within the predetermined distance of the location and determining that the versions of the applications among the plurality of vehicles are different, adjusting a version of the application installed in one or more of the plurality of vehicles.

Systems and methods of coordinating application versions among connected vehicles

A message can be received from a first object. The message can include information about a position of the first object. An electromagnetic energy can be caused to be transmitted in a direction of the first object at a time in which the first object is in motion. A reflection of the electromagnetic energy can be received from the direction of the first object. A first possible position of a second object can be determined based on the reflection of the electromagnetic energy having traveled entirely along a path defined by a line formed by the first object and the vehicle. A second possible position of the second object can be determined based on the reflection of the electromagnetic energy having traveled along a path outside of the line. An actual position of the second object being the second possible position of the second object can be determined.

Detecting an object outside of a field of view

System, methods, and other embodiments described herein relate to improving right-of-way determinations at an intersection. In one embodiment, a method includes acquire, in a lead vehicle that is a cloud leader of a micro-cloud, observations about the intersection for a set of vehicles including at least one remote vehicle. The set of vehicles are approaching the intersection. The remote vehicle and the lead vehicle are members of the micro-cloud. The method includes deriving an assignment of right-of-ways indicating an order about how the set of vehicles may proceed through the intersection. The method includes providing the assignment to at least the remote vehicle to control right-of-way at the intersection.

Systems and methods for collaborative intersection management

System, methods, and other embodiments described herein relate to resolving a standoff by a vehicle. In one embodiment, a method includes generating a happens-before relationship that explains events between the vehicle and other vehicles before the standoff. The standoff may be a dispute for a right of way between the vehicle and the other vehicles. The method also includes identifying the standoff using a causality relationship analysis according to the happens-before relationship. The method also includes generating a mitigation plan for the standoff that forms standoff solutions in association with the standoff being similar to a prior standoff. The method also includes resolving the standoff by causing vehicle maneuvers associated with the vehicle according to the standoff solutions.

Systems and methods for vehicles resolving a standoff

A system, method, and non-transitory computer-readable medium allows for value-anticipating task offloading. The system may include one or more processors and a memory having a task manager module. The task manager module causes the one or more processors to receive a task identifier of a computational task for an application being utilized by a vehicle processor of a vehicle and a state vector describing at least one state of the vehicle and determine, using a utility function, a utility score of the computational task using the task identifier and the state vector which represents an improvement in a functioning of the application if the computational task is offloaded to an external system for processing. Based on the utility score, the one or more processors may offload the computational task to the external system, process the computational task by the vehicle processor of the vehicle, or discard the computational task.

Growing community of inventors

Mountain View, CA, United States of America

Chang-Heng Wang