Ryosuke Tasaki

WiFi RSS fingerprint database construction for mobile robot indoor positioning system

Mobile robot positioning in an indoor environment via fingerprinting technique is made by matching the unknown WiFi data to a spatial WiFi power map database. In past years, this technique has gained reasonable accuracies in the cost of high labor yielded for the database. Thus, automatic database construction by means of interpolating missing data is desired. This paper described a variation of Inverse Distance Weight (IDW) interpolation which is the Modified Shepards Method (MSM) for our medical-oriented mobile robot - Terapio. By properly selecting the reference locations, we found that this method is better than the conventional IDW method and comparable to the recently popular Kriging interpolation algorithm in an indoor environment by employing the state-of-the-art WKNN positioning algorithm.

Adaptive input shaping control of a rotary hook

In crane systems, rotary hook plays an important role in adjusting the skew angle of payload to match with skew angle of vessels or container trucks. At the present, rotary hook is mainly controlled manually by experienced and skilled workers. Therefore, this paper presents to develop an adaptive input shaping (AIS) algorithm for rotary hook system which can automatically drive the payload to desired angle while suppressing the residual oscillation caused by flexible ropes. An algebraic estimator will be the center of adaptive input shaping scheme. Moreover, the comparison between AIS and built-in robust Specified Insensitivity (SI) input shaper will be also conducted when both of them are shaped to be robust with the natural frequency uncertainty of rotary hook system. All of conclusions will be verified through simulation results.

Visual Nonlinear Feedback Control of Liquid Level in Mold Sprue Cup by Cascade System with Flow Rate Control for Tilting-Ladle-Type Automatic Pouring System

This paper presents the novel approach of liquid level control of mold sprue cup using camera information for tilting-ladle-type automatic pouring system in casting industry. A liquid level in mold sprue cup should be kept at high position to avoid inclusion of oxides and dusts during pouring process. A cascade control system with double feedback loop to enhance control response was proposed for the present level control: inner loop was the flowrate control system for pouring ladle, and outer loop was the level control one with camera measurement to detect the liquid level. The control part of flow rate was constructed based on the past study. The mathematical model of liquid behavior in a sprue cup of an inverted cone shape was built in this paper. The level controller was newly designed based on nonlinear control theory by exact linearization. The effectiveness of the proposed cascade system was verified by control simulations.

Pressing Control Design Based on Analysis of Metal Temperature and Flow in Sand Mold Press Casting

The sand mold press casting method is a novel iron casting process, which has developed by our group in recent years. The proposed method can cast iron into complex shapes with a high yield over 90% and produces high-quality products by filling control of molten metal during pressing motion. However, molten metal inside mold is cooled down by heat transfer to the molds and atmosphere, and often causes solidification before the end of press. Then, the pressing movement of the upper mold is blocked by the solidified metal. Therefore, to avoid the solidification during pressing sand mold, metal temperature must be heated properly to estimate the filling temperature decrease behavior based on analysis results of CFD simulator, FLOW-3D. The necessary condition not to solidify molten metal before the end of press has been found. It was made clear that the mold must be heated up to the necessary temperature beforehand in the pressing process. Furthermore, optimum velocity reference with specified pressure constraints has been designed to prevent casting defect such as penetration and also to minimize the temperature decrease. In this paper, optimum velocity control of servo cylinder considering the both of allowable pressure of molten metal and starting time of solidification is proposed by using a theoretical approach of Model Predictive Control: MPC method. The effectiveness of the proposed control system has been demonstrated by computer simulation and experiments using a laboratory scale machine with molten metal of casting iron.

Medical round robot tracking a specified person based on highly accurate position estimation considering attitude angle compensation of wearable sensor

We have developed an innovative medical round robot “Terapio” in hospitals. This novel robot is mainly tasked to deliver medical armamentarium and manage round data. An implemented omni-directional mobile mechanism and a person tracking control system to follow a medical doctor realize smooth transferring movement from a nurse station to a patients bedroom. In this paper, person tracking control using both Laser Range Finders(LRFs) and inertial sensor is proposed to avoid occlusion problem in hidden space of target person and also not to see the target among many passengers. Then, high accuracy position estimation by compensated inertial sensor data considering attitude angle information has been investigated. Person tracking experiments in an occlusion environment and in some passengers have been done, and then the effectiveness of the proposed method has been demonstrated.

Motion control of novel power assist lift robot integrated with omnidirectional assist vehicle considering suppression of limit cycle at grounding

In this paper, system design and motion control of a nursing lift robot which runs on the ground is presented. This supporting lift robot provides a power assist function for transfer aid system with hoisting and lowering, and also an omnidirectional motion function, using DDSS (Differential Drive Steering System). Further, logic control to suppress limit cycle of lift at grounding is proposed. The effectiveness of the proposed robot has been demonstrated through a lot of experiments in the present research.

Dynamic grinding manipulation based on removal volume rate control

We have developed a grinding motion control technology for an automatic finishing process by newly constructing a feedforward driving-type end-effector module attached to a multi-axis industrial arm. Metal products formed by pressurized filling methods, such as vacuum or die casting processes, often have physical penetration defects on the product surface. However, the manufacture of products with complex shapes that is an advantage of these casting methods would be a disadvantage in view of the difficulty of performing the finishing process using a grinding tool. This research is undertaken with a view to realizing an automatic high-speed grinding process for complex-shaped iron products by introducing a macro-micro dynamic control method that simultaneously manipulates a multi-axis arm for global motion and an end-effector module for local motion. Automatic grinding motion control for shorter process time and the associated movement velocity design methods, considering maximization of grinded volume rate per second, are discussed. The newly developed end-effector module, to which a high-speed driving actuator and a high-accuracy force sensor are attached, can detect position and height of a defect in real-time and adaptively moves to grind each defect shape for a required surface quality in a short time.

Robust Sliding Mode Control with Integral Sliding surface of an underactuated rotary hook system

In this paper, robust Integral Sliding Mode Control (ISMC) will be proposed for an underactuated rotary hook system in order to deal with parametric uncertainties presenting in the system parameters. By using the proposed ISMC, the reaching phase encountered in the conventional SMC will be eliminated and as a consequence, the dynamic ambiguity, which occurs when the system is in the transient state forwarding to the sliding mode, can be completely avoided. Furthermore, when the system slides on the sliding surface, the constraints for controller gains will be clarified in order to ensure the robust stability of the closed-loop system. All of conclusions will be drawn through both simulation and experimental results.

Moving target localization method using foot mounted acceleration sensor for autonomous following robot

The development of an autonomous mobile robot system has expanded its application across various fields such as for medical service support, automatic vehicle operation, and delivery businesses. In recent years, we have developed a mobile robot which is tasked to follow a medical doctor from an arbitrary starting point to a patients room, manage medical resources and a patients electronic information in a hospital. The robot follows with its position being just a few meters behind the moving target person or the medical doctor. Here, when there is a blocking object between the target being followed and the observation sensor, the robot stops because it loses the targets position. Obstacles for these robots can be passers-by in straight corridors and corner walls encountered during turns. In this research, we propose a moving target localization method that directly measures or indirectly estimates the position of the robots following target. A wearable acceleration sensor is attached to one-leg of the following target person and self-localization with high accuracy can be estimated by our proposed calculation algorithm based on inertial navigation information. As a compensation for positioning errors accumulated over time, the estimation calculation is processed according to correction information of the stance phase judgment by an acceleration sensor and the targets moving direction by LRF during walking in a straight line. Construction of the estimation compensation algorithm, its implementation on the mobile robot and a trial experiment of the robot following a specified person are described in detail.

Prediction control for high speed positioning with vibration suppression of paper streamer processes using stepping motor

This paper presents a predictive control system of stepping motor to realize the high speed positioning, and the residual vibration suppression by changing pulse input timing suitably. For the only four pulse input, high speed positioning without residual vibration is strongly required in the high-speed mounting and transferring process of semiconductor tip. In this process, it was made clear from our present experiments that the angular variation response is related to the frictional load between the transferred tape and the guide rail part. The load torque for stepping motor is then added to driving system and the influence of load torqe variation on the angular response variation is studied. Here, we investigate the relationship between angular response characteristics, the load torque and the pulse timing, and propose the predictive control system of stepping motor by feedback of one data in the transient states.