Yutaka Uchimura

Bilateral robot system on the real-time network structure

This paper presents a bilateral robot system, which is driven by the static friction-free drive system and implemented on the real-time network structure. The goal is to realize a force reflecting bilateral teleoperation with haptic impression transmission over computer networks. The paper considers two subjects relating to the bilateral robot. The first is static friction, which degrades the performance of manipulation and results in a poor haptic impression. A new transmission mechanism named twin drive system developed by the authors resolves this problem. The transmission mechanism, which resembles the differential gear of automobiles, is essentially free of static friction. This static-friction-free motion greatly contributes to the broad range of motion control applications. The second subject is the time delay of the network, which may cause serious problems such as instability of the feedback system. To avoid such delay, the authors developed a new real-time network protocol stack (RTNP). The detailed mechanism of the twin drive system and architecture of the RTNP are presented, and the control scheme and experimental results are also shown.

Time synchronized wireless sensor network and its application to building vibration measurement

Network based wireless sensing has become an important area of research and various new applications for remote sensing are expected to emerge. One of the promising applications is structural health monitoring of building or civil engineering structure and it often requires vibration measurement. For the vibration measurement via wireless network, time synchronization is indispensable. In this paper, we introduce a newly developed time synchronized wireless sensor network system. The system employs IEEE 802.11 standard based TSF counter and sends the measured data with the counter value. It enables consistency on common clock among different wireless nodes. We describe the accuracy evaluation by simulation studies when the size of nodes increased. The hardware and software specifications of the developed wireless sensing system are shown. The experiments were conducted in a three-story reinforced concrete building and results showed the system performs more than sufficiently.

Wireless network based control with time varying delay

This paper describes a control scheme for a wireless networked system with variable time delay. Especially, the paper focuses on a packet based network in which subsequent packets catch up the preceding packet. The paper reveals that the forward delay and the backward delay, which are the delay from the local to the remote system and vice versa, are needed for the precise system identification and the control method. If only the round trip time is used, the identification and the control for time varying delay lead unexpected results. To measure the time delay, time synchronization of the local and remote system is required and this paper proposes to use a timing synchronization function (TSF) of IEEE 802.11 based wireless network Performance evaluation by simulation studies for system identification and delay compensated controller are shown. The simulation results verified that the proposed method effectively performed.

Controller design based on complete quadratic Lyapunov-Krasovskii functional for time delay systems

This paper describes a controller design method for time delay systems. Conventional design methods based on the Lyapunov-Krasovskii stability theorem may be conservative because they choose the Lyapunov functional to achieve a sufficient condition. In addition, conventional ℌ∞ performance design methods do not focus on the frequency property of a closed loop system; hence, it is difficult to obtain a controller that achieves the desired performance. This paper proposes a stabilizing condition based on a complete quadratic Lyapunov-Krasovskii functional and a controller design considering frequency-dependent performance.

Controller design based on sum-of-squares for time-varying delay systems

This paper proposes a controller design based on sum-of-squares (SOS) method for time-varying delay systems. The SOS condition for controller design and the proof based on the Lyapunov-Krasovskii stability theorem are shown. The performance of the controller is fine-tuned taking into account input-output property. The controller is solved by decomposing a SOS condition, which is given by convex form. Experimental results on a real plant are also described to evaluate the performance of the proposed method.

Mobile robot deployment based on Voronoi diagram

This paper describes a new method for deployment of wireless relay nodes. When applying rescue robots in building or underground city, the wireless radio is attenuated significantly, thus multi-hop extension by relay of wireless communication is required. The goal of research is to deploy relay wireless nodes to maintain connectivity from the base station to the leader robot which explores around the front line. To move the relay robot autonomously, distributed algorithm is required. The method overcomes conventional distributed algorithm, when applying it for wireless relay purpose. In the method, a virtual force which drives a node to the centroid of Voronoi neighbors and it works to maintain the connectivity of wireless communication. The evaluation of proposed method is conducted by numerical simulations and experiments. In the simulation, one or two leader robots are assumed. In the experiment, a mobile robot equipped with omni-wheels is used.

Deployment control of mobile robots for wireless network relay based on received signal strength

Remote operation of robots for rescue purpose requires wireless network to transmit control data and video data, however the communication infrastructure cannot always be used in the case of disaster. Thus the self-organized network needs to be constructed which consequently requires multi-hop relayed ad-hoc network. Because the robot will be used inside the building where the direct wireless communication between the base station and the leader robot. This paper propose a control scheme which aims at the optimum deployment of multi-relay robots using virtual force estimated by the Bayesian filtering based on the state of communication from the received signal strength and location of each robots. Detail of the scheme and simulation results are shown. Experimental results which are conducted with real mobile robots are also shown.

3DZMP-based control of a humanoid robot with reaction forces at 3-dimensional contact points

Expectations for the practical use of service robots have increase of the acceleration of demographic aging. A humanoid robot is a promising form factor for service robots. When a humanoid robot is used in human environments, the robot must be designed considering various reaction forces that act on it. To date most walking robots use the conventional ZMP method for control in order to maintain a stable walking posture. However, the conventional ZMP method cannot handle the various reaction forces that act on contact points in three-dimensional space between a humanoid robot and the environment. To overcome this, a novel control method is proposed, comprising 3DZMP and pZMP. 3DZMP is defined as a point in three-dimensional space where the moment around all axes is zero. 3DZMP can prevent the rotation of a humanoid robot. pZMP is defined as a point where the 3DZMP is orthographically projected on a plane. pZMP is one way to evaluate the stability of 3DZMP. We implemented the proposed method with a prototype and verified that the robot could react to reaction forces, which the conventional ZMP method could not handle.

Real-time network system by responsive processor and its application to bilateral robot control

This paper presents a new network device, which is named responsive processor, plus its message handler, which is a basic protocol for the device. These are designed for a hard real-time control application such as a robot control. By using the responsive processor and message handler, we developed a bilateral robot system that transfers haptic impression over a network. First, we discuss the network system requirements from the view of real-time control. Then, we introduce the main features and functions of the responsive processor. Implementation of the message handler, which is designed to maintain real-time performance, is described in detail. Benchmarks for using the message handler with the responsive processor are also shown. The results of experiments conducted on a bilateral robot system verified real-time performance of the entire system.

A robust force control based on continuous switching of local control rules

A robust force controller based on dual-mode control by continuous switching is described. This controller is applied to an observer-based acceleration controller, which eliminates the disturbance imposed on a manipulator. The robustness of the controller against stiffness variation of objects is improved, and the asymptotic stability of the controller is proved using the Lyapunov function. Experimental results verify the validity of proposed method.<<ETX>>