Search

Detecting and understanding the world is an important step for any intelligent system, but also one that often requires significant processing power. This places great emphasis on the creation of efficient but flexible data processing pipelines. In this paper, we discuss the use of OpenRTM-aist and a component-based approach to rapidly create sensor processing pipelines. The sensor pipeline is made up of a series of filtering and processing steps, provided by pre-existing components for each step. We illustrate the techniques with a pipeline for processing point clouds using the PCL library. [ABSTRACT FROM PUBLISHER]

This paper proposes the task management system to decompose service request from a user into detailed robotic task plans. User's request should be simple, such as “Bring me the newspaper”. However, robotic task plan is complicated and needs to be dynamically generated depending on actual environment and system status. We propose robot profile with capability properties to select robot depending on the task requirement. We developed the task planning method to generate task plan for each robot to realize service execution. Experiment shows that our system is able to realize complex service based on simple command from the user. [ABSTRACT FROM PUBLISHER]

This paper gives an overview of UNR-PF, an open-source, standards-based platform to enable multi-location daily life support services using abstracted functionality from a cloud of robots. Such services require seamless ability-adapted physical support and interaction across multiple areas, provided through collaboration of multiple robots, ubiquitous sensors and smartphones over a network. UNR-PF provides a common infrastructure that separates development of service applications and hardware-based components and intelligently mediates between them. First, we summarize the requirements and design goals initiated in previous work and then describe how the platform was integrated by outlining the implementation of the internal modules and APIs. Finally, we explain how to use UNR-PF by the example of a simple life support scenario and discuss practical aspects of component and service development for integration of a robotic service. [ABSTRACT FROM PUBLISHER]

Cloud computing and smartphone technologies are becoming pervasive in daily life. Cloud robotics, which apply such technologies to robots, is capturing the spotlight as an attractive implementation example. However, when implementing a real time robot service on a wide-range network, it is desirable to grasp the influence of network delay on the robot's actions and the user's experience in advance. Therefore, we propose a simulation environment for Cloud computing robots. This simulation environment replicates the WAN environment between the robot and the Cloud server, and allows the behavior of the Cloud computing robot to be observed in real time. We prototype and evaluate the performance of the proposed simulation environment. [ABSTRACT FROM PUBLISHER]

Collaboration of robots and smart environments is an essential issue when we want to enable a robotic mediator to provide the users with proactive context-aware services in smart environments. However, existing network robot software platforms are not flexible enough to effectively integrate both the respective features specific to robots and to environments. In addition, they do not support high-level collaboration model that is crucial to enable mediating between the user and the smart environment. This paper proposes ICARS, an integrated control architecture for the robotic mediator in smart environments. ICARS consists of three layers each of which provides a flexible communication/device model, an adaptive service model for the integrated robot control architecture and a behavior-based high-level collaboration model as a key feature respectively. We present the detailed design, implementation, and experiments of ICARS. The experimental results show that ICARS enables flexible integration of the diverse devices in the robot and environment, adaptive service provision for collaborative services and easier development of high-level collaborative applications. [ABSTRACT FROM PUBLISHER]

To operate actuators (e.g., robots) with a certain controller in smart environment, it requires some programming for connecting between them. However, such programming without any structure often causes the system with the lack of extensibility and flexibility. In this paper, we show an implementation of the programming framework, called Squarekno t, for connecting between controllers and actuators based on MAC (Mapping between Actuators and Controllers) model. It provides the modularized structure to add new actuators and controllers in the environment easily. It also gives flexibility in customizing the operability for the actuators with the con troller. [ABSTRACT FROM PUBLISHER]