International Business Machines Corporation

Au Optronics Corporation

Sanehiro Furuichi

Masami Tada

Takahito Tashiro

Yasutaka Nishimura

Kenichi Takasaki

Shoichiro Watanabe

Norie Iwasaki

Gaku Yamamoto

Akira Saito

Kazuhito Akiyama

Taku Aratsu

Tomohiro Shioya

Susumu Shimotono

Mari Abe Fukuda

Akira Ohkado

Taku Sasaki

Hayato Uenohara

Satoshi Yokoyama

Hiroya Ogihara

Masana Murase

Asuka Unno

Moriyoshi Ohara

Toru Aihara

Takuya Mishina

Hiroaki Uetsuki

Yohkichi Sasatani

Yuto Fukui

Yasuhide Niimura

Hideo Ichinomiya

Kei Kawase

Hayato Kiriyama

Kiriyama Hayato

Takenori Kohda

Tasuku Otani

Atsumi Ikebe

Hiroshi Tanaka

Hiroshi Ishikawa

Kenji Uchida

Daisuke Maruyama

Noboru Kamijo

Michiharu Kudo

Kazuo Sekiya

Hirotaka Matsumoto

Fumiko Akiyama

Yukihiko Sohda

Akihiro Ogura

Tsend Ochir Bat Ulzii

Satoshi Kawase

Tetsuya Noguchi

Jun Sugiyama

Satoko Tonegawa

Hassan Hajji

Yuka Sasaki

Mitsuru Chinen

Yuriko Sawatani

Reiya Takemura

Shoichiro Wantanabe

Tomoaki Kohno

Aratsu Taku

Taskuku Otani

Tada Masami

Yuriko Kanai

Akiro Saito

A computer-implemented method, a computer program product, and a computer system for parallel cross validation in collaborative machine learning. A server groups local models into groups. In each group, each local device uses its local data to validate accuracies of the local models and sends a validation result to a group leader or the server. The group leader or the server selects groups whose variances of the accuracies are not below a predetermined variance threshold. In each selected group, the group leader or the server compares an accuracy of each local model with an average value of the accuracies and randomly selects one or more local models whose accuracies do not exceed a predetermined accuracy threshold. The server obtains weight parameters of selected local models and updates the global model based on the weight parameters.

Parallel cross validation in collaborative machine learning

A method for calculating traffic flow changes includes detecting a traffic event. The method further includes determining an affected area of the traffic event and determining an investigation area based on the affected area. The method further includes selecting at least one vehicle located within the investigation area and calculating a change in traffic flow due to the traffic event based on a comparison of a predicted traffic flow with a current traffic flow, wherein the current traffic flow is based on information received from the at least one vehicle. The method further includes updating the affected area based on the change in traffic flow and calculating an updated change in traffic flow based on the updated affected area when the updated affected area is larger than a predetermined threshold area.

Calculating traffic flow changes due to traffic events

In an approach for reducing economic loss due to traffic events, a processor receives a set of transportation operation information from one or more transportation service providers about a traffic event. A processor analyzes the set of transportation operation information to identify an area affected by the traffic event and one or more users affected by the traffic event. A processor gathers a set of information about each user of the one or more users affected by the traffic event. A processor prioritizes the one or more users affected by the traffic event. A processor determines an alternative plan for each user of the one or more users affected by the traffic event based on the set of information and a priority number given to each user. A processor proposes the alternative plan to each user of the one or more users affected by the traffic event.

Reducing economic loss due to traffic incidents by proposing actions

A computer system trains a federated learning model. A federated learning model is distributed to a plurality of computing nodes, each having a set of local training data comprising labeled data samples. Statistical data is received from each computing node that indicates the node's count of data samples for each label, and is analyzed to identify one or more computing nodes having local training data in which a label category is underrepresented beyond a threshold value with respect to data samples. Additional data samples labeled with the underrepresented labels are provided, and the computing nodes perform training. Results of training are received and are processed to generate a trained global model. Embodiments of the present invention further include a method and program product for training a federated learning model in substantially the same manner described above.

Training federated learning models

Provided are a computer program product, system, and method for configuring and controlling an automated vehicle to perform user specified operations. User vehicle control programs are loaded in an unmanned vehicle to control the unmanned vehicle to perform a user specified operation. The loading the user vehicle control programs replaces base vehicle control programs in the unmanned vehicle. There is communication with the unmanned vehicle to execute the user vehicle control programs to control the unmanned vehicle to perform the user specified operation. The base vehicle control programs are loaded into the unmanned vehicle to replace the user vehicle control programs to return control of the unmanned vehicle to a vehicle provider after performing the user specified operation.

Configuring and controlling an automated vehicle to perform user specified operations

Computer-implemented methods, systems and computer program products leveraging collection and analysis of anonymized biological data, location data, individual IDs and time data from groups of individuals within a surrounding environment. The anonymized data can be combined with sources of map data and available historical data to help provide context about the surrounding environment of the users and stored for analysis and decision-making that physically impacts and alters the surrounding environment. At periodic or sporadic intervals, the collected data is extracted and analyzed. Based on the analysis of the anonymized data, physical changes are dynamically implemented within the physical environment, including remotely altering the physical environment by instructing changes to surrounding environment over a computer network such as modifying one or more settings of IoT devices positioned within the surrounding environment analyzed.

Dynamic environmental control

The method provides for one or more processors to receive traffic information and passing vehicle information associated with a portion of a roadway in which a passing vehicle approaches and travels through the portion of the roadway. The one or more processors predict travel positions of passing vehicles, based on the traffic information and passing vehicle information. The one or more processors determine an impassible space within an existing lane of the roadway and create virtual lane definitions based on the predicting and the traffic information, in which the lane definitions include an optimum number of lanes, a width of respective lanes, and a lane type, and the one or more processors transmit the lane definitions to the passing vehicles based on a correspondence between a type and width of a respective vehicle and the type and width of respective lane definitions.

Dynamic lane adjustment for multipurpose roadway zones

Provided are techniques for organizing a temporary device group for collaborative computing. A list of functions for each of a plurality of devices are stored. A determination is made to form a device group including a receiver device from the plurality of devices, where the receiver device will perform one of the functions. An individual function usage score is generated. In response to the individual function usage score exceeding a device threshold, a request to form a device group is sent to the receiver device. In response to receiving an indication of acceptance to form the device group from the receiving device, the device group is formed for a temporary period, where functions and data are shared in the device group.

Organizing a temporary device group for collaborative computing

Aspects of the present invention disclose a method for assisting a stalled autonomous vehicle in an incommunicable area. The method includes one or more processors determining a route of an autonomous vehicle. The method further includes identifying an area of the route of the autonomous vehicle that includes a communication outage. The method further includes determining a driving difficulty of one or more segments of the route of the autonomous vehicle within the area. The method further includes determining a location of the autonomous vehicle within a segment of the one or more segments of the route based at least in part on the driving difficulty of the segment.

Stalled self-driving vehicle rescue system

A method and a system for event validation on a vehicle. The method includes detecting an event determined by the vehicle. The method can include requesting and receiving external event determination, from at least one external source within a first surrounding area, based on the event. The method also includes generating an internal event determination based on the event and validating a validated event based on the both the external and internal event determinations. The method includes requesting and receiving external drive plan, from the external source within a second surrounding area, based on the validated event. The method also includes generating an internal drive plan, validating a validated drive plan based on the internal and external drive plans, and implementing the validated drive plan.

Event validation using multiple sources

Adjusting inconsistencies and inaccuracies in location perception, one or more computer processors identify data indicating a location of a computing device; detect a subsequent computing device within a threshold proximity to the computing device; receive data indicating a location of the subsequent computing device; determine a first location reliability score for the identified data indicating the location of a computing device and a second location reliability score for the received data indicating the location of the subsequent computing device; calculate one or more location corrective parameters for the computing device based, at least in part, on the identified data indicating the location of the computing device, the received data indicating the location of the subsequent computing device, the first location reliability score, and the second location reliability score; adjust the data indicating the location of the computing device based on the calculated one or more location corrective parameters.

Location correction utilizing vehicle communication networks

An embodiment of the invention may include a method, computer program product and computer system for managing mobile objects. The embodiment may manage, by a first computing system, a plurality of mobile objects moving within a geographic space. Managing the plurality of mobile objects may assisting with movement of the plurality of mobile objects. The embodiment may determine whether a first mobile object among the plurality of mobile objects is a real mobile object based on a first sensor information received from the first mobile object. The embodiment may use information received from the first mobile object in managing the plurality of mobile objects moving within the geographic space based on determining that the first mobile object is the real mobile object.

Management of mobile objects

Described are techniques for acquiring geospatial data according to an information value. The techniques including determining a context for geospatial data to be used in an application, where the context is based on one or more external factors that influence variation of the geospatial data. The techniques further include calculating an information value of the geospatial data in the context for each of a plurality of information acquisition methods, where the plurality of information acquisition methods include respective data acquisition frequencies and respective spatial resolutions, and where the information value is based on an information loss function, an information amount, and a cost. The techniques further include selecting a first information acquisition method with a highest information value and acquiring the geospatial data using the first information acquisition method.

Geospatial data acquisition based on information value

Embodiments of the present invention provide a computer system, a computer program product, and a method that comprises in response to receiving a data packet from a computing device, classifying the data packet as a task having one or more portions; allocating the classified task to a processing location within a data region based on a location of the computing device; in response to a change associated with the task, dynamically calculating alternate processing locations within a radius of the data region to process one or more portions of the task based on scoring values associated with the change; and redistributing at least one portion of the classified task according to an alternate processing location of dynamically calculated alternate processing locations.

Dispatching tasks and data using multi-access edge computing

A computer system provides navigation services. Route data is processed to extract a plurality of routes, wherein each route comprises a plurality of links and nodes. A partition layer is generated for a map including the plurality of links and nodes of the routes, wherein the partition layer comprises a plurality of meshes each containing links and nodes of the map. A context is processed to remove invalid links and nodes from the map based on the context. The partition layer is updated by merging a subset of the plurality of meshes in response to the removal of the invalid links and nodes. A navigation service is provided for the plurality of routes based on the updated partition layer of the map. Embodiments of the present invention further include a method and program product for providing navigation services in substantially the same manner described above.

Providing navigation services using context-aware trajectory analysis

Growing community of inventors

Tokyo, Japan

Sanehiro Furuichi