Lujia Wang

Towards real-time multi-sensor information retrieval in Cloud Robotic System

Cloud Robotics is currently driving interest in both academia and industry. It allows different types of robots to share information and develop new skills even without specific sensors. They can also perform intensive tasks by combining multiple robots with a cooperative manner. Multi-sensor data retrieval is one of the fundamental tasks for resource sharing demanded by Cloud Robotic system. However, many technical challenges persist, for example Multi-Sensor Data Retrieval (MSDR) is particularly difficult when Cloud Cluster Hosts accommodate unpredictable data requested by multi robots in parallel. Moreover, the synchronization of multi-sensor data mostly requires near real-time response of different message types. In this paper, we describe a MSDR framework which is comprised of priority scheduling method and buffer management scheme. It is validated by assessing the quality of service (QoS) model in the sense of facilitating data retrieval management. Experiments show that the proposed framework achieves better performance in typical Cloud Robotics scenarios.

Real-Time Multisensor Data Retrieval for Cloud Robotic Systems

Cloud technology elevates the potential of robotics with which robots possessing various capabilities and resources may share data and combine new skills through cooperation. With multiple robots, a cloud robotic system enables intensive and complicated tasks to be carried out in an optimal and cooperative manner. Multisensor data retrieval (MSDR) is one of the key fundamental tasks to share the resources. Having attracted wide attention, MSDR is facing severe technical challenges. For example, MSDR is particularly difficult when cloud cluster hosts accommodate unpredictable data requests triggered by multiple robots operating in parallel. In these cases, near real-time responses are essential while addressing the problem of the synchronization of multisensor data simultaneously. In this paper, we present a framework targeting near real-time MSDR, which grants asynchronous access to the cloud from the robots. We propose a market-based management strategy for efficient data retrieval. It is validated by assessing several quality-of-service (QoS) criteria, with emphasis on facilitating data retrieval in near real-time. Experimental results indicate that the MSDR framework is able to achieve excellent performance under the proposed management strategy in typical cloud robotic scenarios.

Towards cloud robotic system: A case study of online co-localization for fair resource competence

The cloud transforms the potential of robotics, which enable poor-equipped robots to fulfill complex tasks. Robots are relieved from hardware limitation, while large amount of available resources and parallel computing capability are available in the “cloud”. We implemented a data management system using Twisted-based server-client platform and Robotic Operating System (ROS), aiming at co-localization of cloud robots. However, resource competition is pervasive for practical applications of networked robotics. As a major bridge, the limited bandwidth becomes a bottleneck needs to be considered for the architecture design. We propose an infrastructure which considers multi-robot autonomous negotiation (MRAN) module. The framework is validated by enabling several poor-equipped robots to retrieve location data from a dynamically updated map which is built by a well-equipped robot. Experiment results demonstrate that the proposed framework is feasible for current robotic applications. Furthermore, it achieves better performance under resource competition, and optimizes Quality of Service (QoS) using a shared network with limited bandwidth.

DP-Fusion: A generic framework for online multi sensor recognition

Multi sensor fusion has been widely used in recognition problems. Most existing works highly depend on the calibration between different sensors, but less on modeling and reasoning of the co-incidence of multiple hints. In this paper, we propose a generic framework for recognition and clustering problem using a non-parametric Dirichlet hierarchical model, named DP-Fusion. It enables online labeling, clustering and recognition of sequential data simultaneously, while considering multiple types of sensor readings. The algorithm is data-driven, which does not depend on priorknowledge of the data structure. The results show the feasibility and reliability against noise data.

A novel radio propagation radiation model for location of the capsule in GI tract

In this paper, we discuss the influence of the antenna orientation radiation pattern in localization algorithm based on Received Signal Strength Indicator (RSSI). We also improve the empirical model of signal propagation by building the path loss function of the human gastro-intestine (GI) tract. The novel model includes information of both the distance and azimuth angle variables. The numerical electromagnetic analysis with the finite-difference time-domain (FDTD) is applied to model the vivo radio propagation channels by using a dipole antenna suitable for the model related to the human body. The proposed propagation model is compared with empirical model, and the simulation results show that the compensated model is more accurate by calculating the azimuth radiation attenuation. It demonstrates that the often overlooked antenna orientation has the dominant effect on the signal strength sensitivity.

Hierarchical auction-based mechanism for real-time resource retrieval in cloud mobile robotic system

In order to share information in the cloud for multi-robot systems, efficient data transmission is essential for real-time operations such as coordinated robotic missions. As a limited resource, bandwidth is ubiquitously required by applications among physical multi-robot systems. In this paper, we proposed a hierarchical auction-based mechanism, namely LQM (Link Quality Matrix)-auction. It consists of multiple procedures, such as hierarchical auction, proxy scheduling. Note that the proposed method is designed for real-time resource retrieval for physical multi-robot systems, instead of simulated virtual agents. We validate the proposed mechanism through real-time experiments. The results show that LQM-auction is suitable for scheduling a group of robots, leading to optimized performance for resource retrieval.

A novel RF-based propagation model with tissue absorption for location of the GI tract

In order to accurately estimate (build) the radio signal propagation attenuation model, especially inside the gastro-intestine (GI) tract of the human body, the Radio Frequency (RF) absorption characterization in human body is investigated. This characterization provides a criterion to design the Received Signal Strength (RSS) based localization system for the objective inside the human body. In this paper, the Specific Absorption Rate (SAR), E-field, H-field of the near and far field are investigated at frequencies of 434MHz, 868MHz, 1.2GHz and 2.4GHz respectively. Then, the numerical electromagnetic analysis with the finite-differencetime-domain (FDTD) is applied to model the in vivo radio propagation channels by using a dipole antenna. Finally, simulation experiments are carried out in homogenous and heterogeneous mediums. The results show that the electromagnetic (EM) propagation is not only distance and orientation dependent, but also tissue absorption dependent in human body. The proposed model is in agreement with measurements in the simulation experiments.

A game theoretical bandwidth allocation mechanism for cloud robotics

Cloud robotics is currently driving interest in both academia and industry, since it would allow robots to off-load computation intensive tasks, combine with multiple robots and even download new skills. Bandwidth allocation is the fundamental and dominant task for resource sharing among users in cloud robotics. However, many technical challenges are still outstanding, since incast congestion happens in high-bandwidth and low-latency networks, when multiple synchronized users send data to a same receiver in parallel [1]. In this paper, we introduce a resource allocation framework for cloud robotics, and propose a game-theoretic problem formulation and linear pricing scheme of bandwidth allocation, we also implement a congestion control algorithm by using optimal parameters derived from the game-theoretic algorithm. Simulation results demonstrate that the proposed mechanism achieves better performance of bandwidth allocation in cloud robotics scenarios.

A Hierarchical Auction-Based Mechanism for Real-Time Resource Allocation in Cloud Robotic Systems

Cloud computing enables users to share computing resources on-demand. The cloud computing framework cannot be directly mapped to cloud robotic systems with ad hoc networks since cloud robotic systems have additional constraints such as limited bandwidth and dynamic structure. However, most multirobotic applications with cooperative control adopt this decentralized approach to avoid a single point of failure. Robots need to continuously update intensive data to execute tasks in a coordinated manner, which implies real-time requirements. Thus, a resource allocation strategy is required, especially in such resource-constrained environments. This paper proposes a hierarchical auction-based mechanism, namely link quality matrix (LQM) auction, which is suitable for ad hoc networks by introducing a link quality indicator. The proposed algorithm produces a fast and robust method that is accurate and scalable. It reduces both global communication and unnecessary repeated computation. The proposed method is designed for firm real-time resource retrieval for physical multirobot systems. A joint surveillance scenario empirically validates the proposed mechanism by assessing several practical metrics. The results show that the proposed LQM auction outperforms state-of-the-art algorithms for resource allocation.

A new calibration method for magnetic sensor array for tracking capsule endoscope

To track the movement of wireless capsule endoscope in the human body, we design a magnetic localization and orientation system. In this system, capsule contains a permanent magnet as the movable object. A wearable magnetic sensor array is arranged out of the human body to capture the magnetic signal. This sensor array is composed of magnetic sensors, Honeywell product HMC1043. The variations of magnet field intensity and direction are related to the capsule position and orientation. Therefore, the 3D localization information and 2D orientation parameters of capsule can be computed based on the captured magnetic signals and by applying an appropriate algorithm. In order to initialize the system and improve the tracking accuracy, we propose a calibration technique based on high-accurate localization equipment, FASTRAK. The calibration method includes two steps. Firstly, we acquire the accurate reference data from FASTRAK tracking equipment, and transform them into the position and orientation parameters of the magnet. Secondly, we calculate three important parameters for the sensor calibration: the sensitivity, the center position, and the orientation. Based on the calibration, we can adjust the magnetic localization and orientation system quickly and accurately. The experimental results prove that the calibration method used in our system can improve the system with satisfactory tracking accuracy.