Abstract: A computer implemented method for configuring a reinforcement learning agent to perform an efficient reinforcement learning procedure, wherein the reinforcement learning agent comprises a model trained using a machine learning process to determine actions to be performed by the reinforcement learning agent. The method comprises using the model to determine an action to perform, based on values of a set of features obtained in an environment; determining, for a first feature in the set of features, a first indication of a relative contribution of the first feature, compared to other features in the set of features, to the determination of the action by the model; and determining a reward to be given to the reinforcement learning agent in response to performing the action, based on the first feature and the first indication.

Abstract: According to one or more embodiments, a node for temporal plan verification and timed game automata, TIGA, controller strategy synthesis for coordination of a plurality of agents is provided. The node includes processing circuitry that is configured to determine a first model of a plurality of temporal plans for timed operations of the plurality of agents violates at least one predefined safety rule, generate the TIGA controller strategy for controlling the timed operations of at least a subset of the plurality of agents where the TIGA controller strategy meets the at least one predefined safety rule and being based on the first model, and cause transmission of the TIGA controller strategy to a dispatch device for translating the TIGA controller strategy to instructions for the plurality of agents.

Abstract: Methods (600) and systems (700) for selecting a frequency band for a device. In one aspect, the method comprises receiving (s602) from the device a request for a particular service type. The method further comprises determining (s604) a set of two or more frequency bands that are available for the device. Said set of two or more frequency bands comprises a first frequency band and a second frequency band. The method further comprises selecting (s606), based on the particular service type being requested and at least one of a knowledge base or a machine learning (ML) model, one of the frequency bands included in the set of two or more frequency bands.

Abstract: It is provided a method for adjusting a network of sensor devices. The network also comprises an aggregation node for aggregating sensor data from at least two of the sensor devices to processed data and an application node configured to approach an application goal based on the processed data. The method is performed in a network device. The method comprises repeatedly performing the steps of: receiving an evaluation of performance of the application node in relation to the application goal; determining that the evaluation indicates that an adjustment is needed; responsive to determining that an adjustment is needed, determining an operation adjustment of at least one of the aggregation node and the sensor devices; and triggering the adjustment.

Abstract: The present disclosure relates to a method of scheduling communication of a wireless communication device in a communications network, and a device performing the method. A method of a device of scheduling communication of a wireless communication device in a communications network is provided. The method includes determining a geographical path along which the wireless communication device is expected to travel, acquiring data indicating power availability of access points via which the wireless communication device is capable of communicating along the expected geographical path, determining, from the acquired data indicating power availability, via which of the access points communication is to be established along the expected geographical path for avoiding establishing communication with one or more of the access points indicated to be subjected to power failure, and scheduling the communication to occur via the determined access points.

Abstract: According to one or more embodiments, a node is provided where the node is configured to: transform a plurality of service level agreement, SLA, values into a plurality of uniform language values associated with at least a first quality of service, QoS, composite metric, and optionally generate a first end-to-end composite SLA based at least on a first set of the plurality of uniform language values.

Abstract: A method performed by a network node for handling one or more operations in a communications network comprising a plurality of computing devices performing one or more tasks. The network node obtains initial parameters relating to the plurality of computing devices, environment and the communications network; and generates a plan by taking one or more operation goals involving the plurality of computing devices into account as well as the obtained initial parameters, wherein the generated plan relates to operation of the plurality of computing devices. The network node further computes a number of back-up plans, wherein the number of back-up plans are taking one or more events into account wherein the one or more events relate to operation of the plurality of computing devices; and executes one or more operations using the generated plan, and in case the one or more events occur, using a computed back-up plan.

Abstract: Robots are deployed for handling different tasks in various field of applications. For the robots to function, task planning is required to be done. During the task planning, goal setting is done, as well as actions to be executed for corresponding to each goal are decided. Traditionally, this is carried out first and then the robots start executing the task plan, thereby failing to capture any change in the environment the robots operate, post the task plan generation. Disclosed herein is a method and system for robotic task planning in which a task plan is generated and is executed. However if the task execution fails due to change in any of the parameters/factors, then the system dynamically invokes an adaptation and re-planning mechanism which either updates the already generated task plan (by capturing the change) or generates a new task plan, which the robot can execute to achieve the goal.

Abstract: This disclosure relates generally to robotic network, and more particularly to a method and system for hierarchical decomposition of tasks and task planning in a robotic network. While a centralized system is used for action planning in a robotic network, any communication network issues can adversely affect working of the robotic network. Further, hardcoding one or more specific tasks to a robot restricts use of the robots irrespective of capabilities of the robots. The robotic agent decomposes a goal assigned to the robot to multiple sub-goals, and for each sub-goal, identifies one or more tasks to be executed/performed by the robot. An action plan is generated based on all such tasks identified, and the robot executes the action plan, in response to the goal assigned to the robot.

Abstract: This disclosure relates generally to human-robot interaction (HRI) to enable a robot to execute tasks that are conveyed in a natural language. The state-of-the-art is unable to capture human intent, implicit assumptions and ambiguities present in the natural language to enable effective robotic task identification. The present disclosure provides accurate task identification using classifiers trained to understand linguistic and semantic variations. A mixed-initiative dialogue is employed to resolve ambiguities and address the dynamic nature of a typical conversation. In accordance with the present disclosure, the dialogues are minimal and directed to the goal to ensure human experience is not degraded. The method of the present disclosure is also implemented in a context sensitive manner to make the task identification effective.

Abstract: Systems and methods for generating control system solutions for robotics environments is provided. The traditional systems and methods provide robotics solutions but specialized to only a particular robotic application, domain, and selected structure.

Abstract: This disclosure relates generally to method and system for evaluating performance of workflow resource patterns. Structured way of resource utilization for performing a complex task thereby evaluating end to end performance in any industrial warehouse is crucial. The proposed disclosure processes the received task to disintegrate into a plurality of sub tasks and for each sub task a workflow resource pattern is generated. Further, dynamically each workflow resource pattern is mapped to corresponding queueing mode of queueing station in the hybrid queueing network model. The system further evaluates end to end performance for each workflow resource pattern by computing performance metrics. Based on the performance evaluation the queuing mode may be replaced with another queueing modes for the workflow resource pattern. The proposed disclosure robustly analyses end to end performance of the workflow resource pattern maximizing throughput, improvising accuracy, increasing efficiency and thereby minimizing cost.

Abstract: This disclosure relates generally to human-robot interaction (HRI) to enable a robot to execute tasks that are conveyed in a natural language. The state-of-the-art is unable to capture human intent, implicit assumptions and ambiguities present in the natural language to enable effective robotic task identification. The present disclosure provides accurate task identification using classifiers trained to understand linguistic and semantic variations. A mixed-initiative dialogue is employed to resolve ambiguities and address the dynamic nature of a typical conversation. In accordance with the present disclosure, the dialogues are minimal and directed to the goal to ensure human experience is not degraded. The method of the present disclosure is also implemented in a context sensitive manner to make the task identification effective.

Abstract: Robots are deployed for handling different tasks in various field of applications. For the robots to function, task planning is required to be done. During the task planning, goal setting is done, as well as actions to be executed for corresponding to each goal are decided. Traditionally, this is carried out first and then the robots start executing the task plan, thereby failing to capture any change in the environment the robots operate, post the task plan generation. Disclosed herein is a method and system for robotic task planning in which a task plan is generated and is executed. However if the task execution fails due to change in any of the parameters/factors, then the system dynamically invokes an adaptation and re-planning mechanism which either updates the already generated task plan (by capturing the change) or generates a new task plan, which the robot can execute to achieve the goal.

Abstract: In order to make use of computational resources available at runtime through fog networked robotics paradigm, it is critical to estimate average performance capacities of deployment hardware that is generally heterogeneous. It is also not feasible to replicate runtime deployment framework, collected sensor data and realistic offloading conditions for robotic environments. In accordance with an embodiment of the present disclosure, computational algorithms are dynamically profiled on a development testbed, combined with benchmarking techniques to estimate compute times over the deployment hardware. Estimation in accordance with the present disclosure is based both on Gustafson's law as well as embedded processor benchmarks. Systems and methods of the present disclosure realistically capture parallel processing, cache capacities and differing processing times across hardware.

Abstract: This disclosure relates generally to method and system for evaluating performance of workflow resource patterns. Structured way of resource utilization for performing a complex task thereby evaluating end to end performance in any industrial warehouse is crucial. The proposed disclosure processes the received task to disintegrate into a plurality of sub tasks and for each sub task a workflow resource pattern is generated. Further, dynamically each workflow resource pattern is mapped to corresponding queueing mode of queueing station in the hybrid queueing network model. The system further evaluates end to end performance for each workflow resource pattern by computing performance metrics. Based on the performance evaluation the queuing mode may be replaced with another queueing modes for the workflow resource pattern. The proposed disclosure robustly analyses end to end performance of the workflow resource pattern maximizing throughput, improvising accuracy, increasing efficiency and thereby minimizing cost.

Abstract: This disclosure relates generally to robotic network, and more particularly to a method and system for hierarchical decomposition of tasks and task planning in a robotic network. While a centralized system is used for action planning in a robotic network, any communication network issues can adversely affect working of the robotic network. Further, hardcoding one or more specific tasks to a robot restricts use of the robots irrespective of capabilities of the robots. The robotic agent decomposes a goal assigned to the robot to multiple sub-goals, and for each sub-goal, identifies one or more tasks to be executed/performed by the robot. An action plan is generated based on all such tasks identified, and the robot executes the action plan, in response to the goal assigned to the robot.

Abstract: Systems and methods for generating control system solutions for robotics environments is provided. The traditional systems and methods provide robotics solutions but specialized to only a particular robotic application, domain, and selected structure.

Abstract: This disclosure relates to managing Fog computations between a coordinating node and Fog nodes. In one embodiment, a method for managing Fog computations includes receiving a task data and a request for allocation of at least a subset of a computational task. The task data includes data subset and task constraints associated with at least the subset of the computational task. The Fog nodes capable of performing the computational task are characterized with node characteristics to obtain resource data associated with the Fog nodes. Based on the task data and the resource data, an optimization model is derived to perform the computational task by the Fog nodes. The optimization model includes node constraints including battery degradation constraint, communication path loss constraint, and heterogeneous computational capacities of Fog nodes. Based on the optimization model, at least the subset of the computational task is offloaded to a set of Fog nodes.

Abstract: This disclosure relates generally to autonomous devices, and more particularly to method and system to optimally allocate warehouse procurement tasks to distributed autonomous devices. The method includes obtaining, at a coordinating agent, a global task associated with the warehouse and information associated with the robotic agents. The information includes a count and status of the robotic agents. The global task is profiled to obtain a set of sub-tasks and constraints associated with the set of sub-tasks are identified. The constraints include utilization constraint and/or pricing constraints. A distributed, decentralized optimal task allocation is performed amongst the robotic agents based on constraints to obtain optimal performance of robotic agents.