The ageing of elderly is a significant problem in developed countries. We address the problem by using a personal robot that follows elderly people. To accomplish its task, the robot needs to meet two main specifications: compact size, and excellent traveling performance. In this paper, we introduce the prototype of our compact personal robot and propose a posture control approach based on a model predictive control. Our approach can control the zero moment point within the desired range to achieve quick turning. The effectiveness of the proposed approach is evaluated by both simulations and experiments using the prototype robot.
The in-wheel motor (IWM) is the most preferred driving mechanism of electric vehicles for its advantages of vehicle motion control, energy efficiency, and vehicle design flexibility. One of the technical problems of the IWM is the reliability of power and the signal wires. Wireless power transfer technology is the best solution to this problem. In this paper, a bidirectional wireless power transfer circuit using a primary inverter and a secondary converter is proposed. We propose a control method for both the inverter and the converter to stabilize the secondary DC-link voltage. The proposed method is verified by simulation and experiment using simulated test equipment.
Time delay problem used to be dealt mainly in process or plant engineering in the early phase of time delay systems research. In recent years, time delay problem in network-based control systems has been attracting much attention. This kind of network is usually unstable, and time delay is time varying and unpredictable. As a time delay compensation method, Smith predictor was a famous and useful approach. However, this method uses time delay model to compensate for time delay effect and so it is not a practical method for network-based control systems. Thus, communication disturbance observer (CDOB) has been proposed to compensate for time delay effect without using a time delay model. However, a large steady state error occurs, if the network is unstable. This paper modifies the structure of a time-delayed control system with CDOB considering an unstable network. To reduce nonlinear effects such as jitter or packet dropouts, the allocation of the system model in the CDOB has been modified. Then, the robustness of the modified CDOB structure against packet dropouts and jitter is confirmed by simulation. The validity of the modified structure is shown by experimental results.
We developed a leg extender exoskeleton that extends the leg length by 40 cm. We also propose a linearity index to evaluate the feeling of operating the exoskeleton. We developed an evaluation method for the foot positioning accuracy of the leg extender exoskeleton. This method uses three types of target marker trajectories to consider various walking situations. We performed an actual experiment with four types of link configurations. The foot positioning accuracy was measured and evaluated according to the linearity index of each exoskeleton. The results showed that an exoskeleton with good linearity provides better foot positioning accuracy.
In this paper, a voltage disturbance observer (VDOB) and counter electromotive force estimation observer (CFOB) are proposed. A voltage disturbance is defined as the electrical disturbance of DC brushed motors. A VDOB estimates a voltage disturbance from a voltage reference and the measured current. Utilizing a VDOB, high-performance motion control is realized. A CFOB estimates the counter electromotive force from an estimated voltage disturbance. By estimating the angular velocity from the counter electromotive force, position-sensorless motion control is realized. Therefore, high-performance position-sensorless motion control of DC motors is realized by combining these observers. The validity is experimentally verified.
The developments in position control methods have enabled us to obtain a rapid response or high control bandwidth for improving control performance. However, there is an emerging problem in which a control system with a high control bandwidth may excite mechanical vibration, which deteriorates control performance. This is the case especially in a mechanism that has a flexible joint between the motor and the load, such as the feeding tables of machine tools, the arms of industrial robots, and rolling mills. Therefore, there is a strong demand for vibration suppression control with a high control bandwidth while maintaining productivity. For high-precision control of a two-inertia system, the position information of both the motor side and load side is usually required to obtain a high control bandwidth. In order to reduce the implementation cost and space, a novel control method, which employs the load-side information only, is proposed using a high-resolution encoder. Simulation and experimental results demonstrate that the proposed method is implementable and exhibits good control performance.
The paper presents a vision-based estimation method for the landing foot angle of a biped walking robot on uneven terrain. The authors previously proposed a vision-based compliant landing method. However, this method only considers flat and horizontal ground, and the landing foot angle estimation by internal sensors such as encoders may cause error to accumulate. To overcome this problem, the present paper proposes directly estimating the landing foot angle by using visual sensing information. Furthermore, stable contact after landing is achieved by reaction force feedback considering the landing angle. The validity of this method was evaluated through several experiments.
High-precision stages require high-speed and high-precision control to improve their production throughput and quality. However, their motion speed and accuracy are expected to reach a limit in the near future if the conventional high-precision stage structure is used. Therefore, the authors designed and fabricated a ‘catapult stage’ which has a structure that can be decoupled into a fine stage and coarse stage. The catapult stage is different from conventional dual stages where the fine stage is disturbed by the coarse stage because they contact each other. This paper proposes a novel control system design for the catapult stage and a control method that shortens the settling time using final state control (FSC). So far, FSC has mainly been used for applications such as hard disk drives for which the initial states are zero. However, it is important to consider the initial states for the catapult stage because the initial position, velocity and acceleration of the catapult stage are not equal to zero. Simulations and experiments were performed to demonstrate the effectiveness of the proposed methods.
Usually, a belt driving system has low resonance because of the elasticity of the timing belts, and the low resonance may degrade the response performance and positioning precision. This paper presents a robot joint that links the motor side to the load side via a timing belt. This can be regarded as a two-inertia system. This paper proposes a novel control method that uses multiple motors on both the motor side and load side of a MISO system. The proposed method achieves vibration suppression and phase stabilization simultaneously. The implementation is relatively simple because the design only requires two parameters. Simulations and experiments were performed to demonstrate the effectiveness of the proposed control system.
This paper proposes a semicircular tubular linear synchronous motor called a “circular shaft motor” (CSM). A CSM can realize the direct-drive motion along the circumference of a circle. The design of the CSM is described. A magnetic field analysis was performed for the CSM model. The thrust characteristics of the prototype CSM system were measured in experiments. Angle control and force control were performed to demonstrate the utility of the developed CSM.
To improve the technology for loss estimation and the efficiency of DC-DC converters, this paper focuses on air-core reactors. First, the author tests a reactor for the increase in temperature by changing the waveforms and frequencies under the same cooling and RMS current conditions. These conditions enable separation of the DC and AC losses. The results for the increase in temperature are different and demonstrate the existence of AC loss in the air-core reactor. It is determined that the AC loss is the eddy current loss by a finite element analysis. Next, this paper calculates the eddy current loss in the DC-DC converter. The ripple current in the DC-DC converter causes no-load loss, which consists of DC and AC losses. The former is the Joule loss due to the ripple current, and the latter is the eddy current loss due to the ripple current. The eddy current loss accounts for more than 90% of the no-load loss. Finally, this paper discusses the reduction of the no-load loss. Electromagnetic coupling is an effective way to reduce the no-load loss. The no-load loss decreases to 23% by electromagnetic coupling.
A time delay between a master robot and slave robot degrades the stability of a bilateral control system. Thus, this paper presents an analysis on the relations between the position controller's gain on the master side, the control performance, and the stability. The analytical results showed that the stability is guaranteed when the gain is equal to the stiffness of the contact object. Accordingly, an adaptive controller that dynamically determines the gain depending on the stiffness was designed. The validity of the adaptive controller was verified through simulations and experiments.
A resonant chopper that operates with a half cycle resonance is used for higher efficiency in DC/DC converters. The on-period of the switch in the resonant circuit is held constant to maintain the soft-switching operation, because it depends on the LC resonance period. In addition, the operating frequency must be reduced to adjust the output power. Thus, we propose a new quasi-resonant chopper that has a second resonant capacitor with the auxiliary short-circuit switch in parallel. The on-period of the chopper is changed by manipulating the short-time of the second resonant capacitor while maintaining a resonant state. As a result, the output power of the resonant chopper can be adjusted with the soft switching operation.
This paper proposes a single-pulse control method for improving the efficiency of switched reluctance motors. The proposed method has three excitation modes to vary the excitation interval according to the conduction angle reference value. Experiments were carried out with a real system to compare the efficiencies of the proposed method and the voltage-PWM control with a constant excitation period. The efficiency of the proposed method was a maximum of 13.2% higher in the low-medium speed and high load regions compared with the voltage-PWM control.
This paper presents a new topology optimization method based on normalized Gaussian network (NGnet). In this method, the machine region is subdivided into small elements whose material states are determined from the output of NGnet so that the objective function is extremized under the given constraints. The present method is applied to shape optimization of the rotor in an interior permanent magnet motor. The rotor shape is optimized to minimize the torque ripple while keeping the average torque. It is shown that the optimization halves the torque ripple while the average torque remains unchanged. This result was validated through a comparison between the computed and measured torques.
This paper proposes a new sensorless vector control method for induction motors using a minimum dimensional flux D-state observer with instantaneous speed estimation. First, the D-state observer with a new design rule of constant observer gains is established in a different way. Second, a new speed estimator using flux estimates produced by the observer is established on the basis of the fundamental relationship between the rotor flux and the slip frequency. Speed estimates are reused in the observer as feedback. Characteristic analyses and the usefulness of the proposed vector control method are verified by simulations.
A switching control strategy for extending the soft-switching range of a three-phase dual-active-bridge (DAB) galvanically isolated bidirectional dc/dc converter under the zero-voltage-switching (ZVS) operating mode is proposed. The converter topology consists of two three-phase active bridges linked by Y-Y connected three-phase high-frequency transformers. One drawback of the conventional phase-shift control strategy is that soft switching is only possible within a restricted converter operating region. A novel pulse-width-modulation strategy to extend the conventional soft-switching operating mode region and its analysis are presented in this paper. A 10-kW prototype has been built, and experimental results are presented in order to demonstrate the validity of the theoretical analysis and the practical feasibility of the proposed strategy.
This paper describes the characteristics of a new multilevel inverter that reduces the number of element in which the current flows through in comparison with a conventional NPC-type inverter. Digital control by the microprocessor was realized by simplification of the control block and the switch of the signal wave. A high efficiency more than 98% was identified by an experiment.