With the rapid spread of smartphones, smartphone addiction has become a problem．In this research, we focus on the physical contact from a wearable robot and aim to create a system for mindset and behavior change about the smartphone usage time of a user. We propose a wearable robot that moves on the user’s arm and alerts the user by squeezing interactions when the user uses the smartphone for a long time. In this paper, we introduce a prototype of the movement mechanism and position estimation system for the wearable robot we developed as an initial step in the goal, and show the results of an evaluation experiment of the position estimation system in a virtual environment.

The final goal of this research is to achieve a kind of manipulation which allows small collisions. The approach proposes dealing with the problems of impact forces in 3 main steps: mechanical device, control, and manipulation strategy. Although there are several control-based research about compliance they all have an intrinsic response time, which might not be ideal under some situations. Thus, a mechanical solution to cover up for the mentioned period is desirable. In this phase of the research, a robot joint which uses compressed air to implement variable stiffness was designed and built under requirements to allow small collisions. The data from stiffness experiments showed that the designed device is close to real implementation. Collision experiments were also performed, and the results show some tendency towards the desirable behavior. However, more tests are necessary for a complete evaluation.

This research aim is to develop a rotary damper with Bingham fluid for robot joints. The SEA method, which incorporates elastic elements in joints, is used in robot joints as a countermeasure against contact and collision. The disadvantage of this method is that the mass-spring system causes vibration due to contact and collision. This study examines the improvement of this drawback by incorporating a rotary damper using a Bingham fluid. In this study, a rotary damper was fabricated, and experiments were conducted to measure the yield torque as the ability to lock the shaft during non-collision, and the damping torque as the ability to dampen as a damper. Both torques were confirmed by the experiments.

In this paper, we described an automation system using a robot arm and machine learning for the press process of the polymer materials development. In the system, the press machine is operated through robot manipulation and control signal from the system. As an evaluation of a molded polymer, we constructed a method to recognize the polymer by image processing and calculate its thickness. Evaluation functions were proposed to evaluate the experimental process, which considers the thickness of molded polymer and press time. We implemented experimental parameter exploration by Bayesian optimization for the next experiment using those evaluation values. Through verification experiments, it was confirmed that the system could proceed the experiment sequentially by proposing press parameters that satisfies the experiment conditions.

In this paper, a flexible laboratory automation system based on a distributed framework that is aiming for autonomous scientific experiments is described. The system is also based on the autonomous experiment loop consisting of sample production, evaluation, and parameter exploration. As an example of applying the system, an implementation for the press process in the development of polymer materials was presented. The system utilizes robot manipulation for sample production, evaluation of the sample, and experimental parameter exploration by Bayesian optimization. The experiment loop was enabled to be conducted using the system.

In order to further advance the research and development of flapping flying robots and to incorporate them into our daily lives, it is necessary to make more researchers and the general public recognized and interested in flapping flying robots. Therefore, it is necessary to search for a safe, simple, and easy-to-understand manufacturing method. In this research, we developed a servo-type flapping flying robot that anyone can easily create so that it can be recognized as more familiar research themes.

We have developed an improved flapping UAV with a payload sufficient to carry a micro-camera system. In order to increase the payload of the airframe, in addition to changing the structure of the struts that make up the airframe, we reviewed the gravity center moving mechanism and the tail structure, which are the characteristic parts of the flapping UAV that we researched and developed and improved the parts and structure. As a result, we succeeded in reducing the weight by 11% compared to conventional UAV. Also, by changing the material of the gearbox, the lift reached 16.3 gf and the UAV successfully flew with a 13.6 g micro camera on board.

This paper describes an unmanned aerial vehicle (UAV) system constructed to patrol river areas and provide evacuation guidance in place of local government officials in the event of a flood. In this system, the unmanned aircraft operates by switching between residual person check and evacuation guidance. In the residual person check, the system detects residual persons using an RGB camera and an infrared camera while flying along the route. For evacuation guidance, the unmanned aircraft tracks the residuals based on the detection results, and supports the evacuation with a speaker. The operation of the constructed system was verified through flight experiments.

The article presents a new aerial robot unit that can perform both aerial assembly and disassembly. The robot is controlled as an under-actuated quadrotor in the unitary state and as a fully-actuated octocopter in the assembly state. The system includes novel docking mechanism, rotor configuration, and control system that enables the transition between two control models during assembly and disassembly motions. We also presents a motion strategy for aerial assembly and disassembly motion that includes a recovery action for hazardous states. The robot’s capabilities were tested through experiments, which demonstrated a success rate of aeiral assembly/disassembly is approximately 90%. Moreover, in the assembly state, the robot can utilize full-pose tracking and generate more than nine times the torque of a single unit. To the best of our knowledge, this is the first robot system that can perform both assembly and disassembly while seamlessly transitioning between fully-actuated and under-actuated models.

Landing on high altitude places is considered to be an effective mean for fixed-point observation and also facilitate work done by UAVs. In this research, we developed a landing mechanism that allows the UAV to land in different types of locations. We previously developed a passively perching device which allowed the UAV to perch on cylindrical structures such as a wooden pillars and also land on a flat surfaces, depending on the contact. Based on this research we improved on the passively perching device with an adaptive mechanism, which allows the UAV to perch on branches. In order to evaluate this device, we demonstrated its usefulness by confirming its range of motion and its ability to land on an actual environment.

Smooth navigation, which is consisted of efficiency and comfort, is important for autonomous mobility scooters. Efficiency can be regarded as the shortness of time the vehicle requires to reach the destination. In a previous work, the theory of situation awareness was applied to the passenger’s comfort recognition. Based on the theory, the passenger was considered to recognize comfort according to not only the recognition of the current situation, for example, the distance between the vehicle and the nearest pedestrian, but also the prediction of the future status. In this paper, passenger’s tolerance was considered as relatively smaller discomfort against close pedestrians. The characteristics of this tolerance was investigated by analyzing data of the previous work and conducting a participant experiment. The results revealed situations that the vehicle can refrain from avoiding approaching pedestrians. Based on the results, design requirements of a path planning method for smooth autonomous navigation were set.

Accurate vehicle position estimation is essential for automated driving and mapping. GNSS (Global Navigation Satellite System), IMU (Inertial Measurement Unit), and wheel speed are used for vehicle position estimation. However, some vehicles have difficulty in acquiring wheel speeds, which limits the range in which they can be used. Therefore, this study proposes a method to improve the accuracy of DR in sections where the FIX solution is not available by estimating the speed using SLAM with LiDAR, which is often used for automatic driving and mapping, with the goal of achieving the same performance as the method with wheel measurements.

In this research, a novel approach to the classical problem of package depalletizing in logistics systems is presented. The conventional depalletizing process relies on rule-based methods to ensure safety, but it only allows for a single type of action. This limitation can make the process slow and inefficient. We propose a hybrid depalletizing algorithm that combines a rule-based mothed with data-driven learning using Q-learning, which allows for shorter trajectory actions. This hybrid approach not only maintains the safety aspect of the conventional approach, but also enhances the efficiency of the depalletizing process. We also validate the algorithm in both simulation and real-world environments with clutters of different sizes. The results show that the algorithm increases efficiency by 19% compared to traditional methods, demonstrating its potential for efficient and safe depalletizing in logistics systems.

We propose an unconventional car equipped with a camera monitor that allows the driver to check safety in all directions simply by looking ahead. One of the features of this system is that all directions, including the front, are replaced by a camera monitor. The effectiveness of this system was verified by creating a car model in a VR space and having the subject drive the car. As a result, the car with 180 °front view and 90 °left, right, and rear view was superior in terms of reaction time, and the car with 180 °front and rear view was the easiest to drive in terms of subjective evaluation.

In many production industries, it is required to develop products with higher-performance in shorter term, because a life cycle of the products is becoming shorter. For such requirement, a methodology of rapid prototyping is introduced. In system developments with electronic components, it is necessary to select components with characteristics suitable for an environment in which the components are applied. For this purpose, it is aimed to construct a support system for comparing sensors in rapid prototyping based on a framework of plug and play connection to automotive network which we have developed. In this paper, GUI tools for a work which converts values acquired from sensors to values in an unit used usualy for distance, temperature and so on is introduced to the framework to reduce costs for the work. Finally, an effectiveness of proposed system in comparing distance sensors is shown through an experiment using real mobile robot.

Trajectory generation is an important technique for making automated driving comfortable and safe. In this paper, we focus on deceleration trajectories due to pauses. Ideally, the deceleration trajectory should change the way the vehicle decelerates according to the surrounding environment. In the past, deceleration trajectories were generated with the goal of decelerating sufficiently before the stop position to stop smoothly, and therefore, no consideration was given to adjusting the degree of deceleration. Therefore, it is difficult to change the way the vehicle decelerates according to the environment. To solve this problem, we propose a method to generate a deceleration trajectory based on a speed profile, which is a target speed set at each point on the route according to the surrounding conditions.

A Real Driving Emission (RDE) test procedure for type approval was started in Japan in 2022. In Japan, test vehicle is driven on a test road (test course) by tracing a speed pattern prepared per RDE test. In the test, the vehicle has need to trace the patterns in the same manner as the conventional chassis dynamometer test. To trace the target vehicle speed at these tests precisely, the use of a driving robot is effective for reproducibility. In this study, we developed the driving robot for the electrified vehicles (HEV, Plug-in HEV, BEV, FCV), which have the regenerative brake control. We applied a new logic to our self-product robot for the engine vehicles. To change the operation timing from the accelerator pedal to the brake pedal, we focused the braking deceleration of the regenerative brake control. At the chassis dynamometer and the test road, we verified the improvement of the speed traceability by the driving robot installed this logic.

This paper evaluates motion planning and automatic piercing control of small vessels considering uncertainty such as the position error caused by wind, waves, currents and the sensor caused by the ship’s motion. Generally when navigating at low speeds, the effects of external disturbances such as wind and waves increase, and manoeuvring the ship in consideration of these factors to avoid collisions is a very heavy load. Therefore, the demand for automatic berthing control for small vessels has increased. In the following discussion, we implement Hybrid A* Algorithm-based motion planning along with the controller characteristics and the effectiveness of the proposed motion planning is validated experimentally.

Nowadays high-definition (HD) map is widely used in autonomous driving to acquire correct information of road during driving. At the same time, HD map also needs to be updated promptly when road changes happen, such as the establish of new traffic signs or the expansion of road lanes. Therefore, developing a system that automatically detects road change areas is an essential issue need to be solved. In this paper, we propose a self-position and pose estimation method using a low-cost drive recorder and millimeter-wave radar by estimating and correcting the bias between the positioning result of MMS-Probe and the true value of HD map.

In this paper, we propose a method for obstacle detection and height estimation using a fish-eye stereo camera. Disparity and intensity information is used. The method using only disparity information may incorrectly detect road surface as obstacles. Therefore, we obtain the intensity of the road surface by using the obstacle information obtained from the disparity and compare the intensity throughout the entire image to detect obstacles. Experiments show that the proposed method using the disparity and intensity information can detect only obstacles without detecting the road surface as obstacles. Experiments also show an improvement in the accuracy of height estimation.

The subject of research is a remote control input device for unmanned aerial vehicles（UAVs). There are various types of the devices for the UAVs, and most of them consist of joysticks, which intrinsically require coordinate transformation. In order to overcome the difficulty, firstly the necessary and expected features of the remote controllers for the UAVs were examined with the upstream design tools, ”Value graph” for example. Based on the design analysis, we proposed a remote control input device that aimed to reduce operator fatigue and to enable intuitive operation. The effectiveness of this proposed device was discussed in this paper with the experimental results.

The behavior of a drone flying in a narrow space change significantly due to the influence of surrounding obstacles, and it is thought that the risk of drone crashes can be reduced by knowing the changes in drone behavior in advance in use cases such as inspections that require the drone to approach obstacles such as walls. In previous research, a test obstacle with walls covering five sides except for the entrance showed that drone behavior is stabilized when an opening of a certain size is provided at the bottom of the surrounding wall [1]. Therefore, with the aim of developing a test obstacle that corresponds to the use case, this study aims to investigate the influence of the wall surface on the drone through experiments using test obstacles to investigate the influence of the wall and ceiling surface and visualization of the flow around the drone through simulation.

A fixed pitch propeller is generally used for a small type electric unmanned aerial vehicle with fixed wings or electric unmanned Vertical Take-Off and Landing (eVTOL). On the other hand, when a variable pitch propeller can be applied, propulsion efficiency is greatly improved, it can be expected that flight performance such as cruising time will be greatly improved. However, at present, there are problems with miniaturization and weight reduction of the variable pitch propeller mechanism, and it is not generally used. We aim to develop newly design technologies to overcome these problems. In this paper, we describe the effect of force acting in the pitch direction of blades with the variable pitch propeller mechanism in rotating on the mechanism stability. It is applied to the electric Quad Tilt Wing VTOL (weight 25 kg) being researched and developed by Shizuoka Prefecture and JAXA.

The shortest path should be desired for considering flight time or battery capacity when the path of a UAV is planned. This study assumes a no-fly zone in a flight area where a UAV cannot travel. Under this condition, we planned the path of a UAV by ant colony optimization in computer simulations, then conducted the auto-flight experiment of the UAV on the obtained path.

This paper reports on the development of an aerial manipulator equipped with pneumatic actuators. By using pneumatic actuators, it is possible to fabricate a lightweight, high-power manipulator with excellent shock resistance and environmental resistance. As a prototype of aerial manipulator, a pneumatically-driven 3-DOF manipulator was designed, fabricated, and mounted on a small quad drone with a wheelbase of 450 mm. The manipulator weighs approximately 610 g, and the total weight of the aerial manipulator is approximately 1.7 kg. Ground experiments of the manipulator are successfully conducted including a 0.5 kg payload operation test and a 200 mm square-drawing test with a marker pen. In the flight experiments, the drawing manipulation test were conducted with a manual attitude control of the drone, which clarifies some technical issues for the aerial manipulator.

In conventional methods of bilateral control-based imitation learning, restrictng a model’s output was infeasible once the model is trained. This led to safety issues, and hindered industrial application of bilateral control-based imitation learning. This study applied a variational autoencoder (VAE) to bilateral-control based imitation learning. A VAE is one of the encoder-decoder models with capability to present its latent variables as their Gaussian distributions. By sampling multiple sets of latent variables from one set of their distributions, generating multiple sets of commands from one set of input is made possible. In the experiments, this study verified constrained motion planning in bilateral control-based imitation learning using a VAE, by first generating command value using the average of derived latent variables, and then repeating attempts of sampling from their distributions, until reaching desired command value.

In recent years, many cases of image discrimination and machine control using deep learning have been reported.

In our previous research, we implemented a neural network on a mobile robot and used color images, distance images, direction data, destination numbers, and route selection numbers five times each to output vehicle speed, parallel movement direction, and rotational speed. bottom. At the same time, in order to detect an unlearned environment and stop it safely, we also established a method for detecting an unlearned environment and forcibly stopping it using an autoencoder that estimates depth images from color images.

In this paper, in order to realize inter-floor movement by elevator, we prototyped an arm with a camera that can operate elevator buttons. Therefore, we designed and learned a neural network that can simultaneously estimate vehicle and arm control parameters, and confirmed its effectiveness through experiments.

In general, a robot dynamics model is derived from the equations of motion, and the dynamics parameters are obtained through identification experiments. However, it is difficult to obtain accurate parameters because of the influence of noise and un-modeled dynamics. These errors may be enhanced by a feedback controller, which degrades control performance. Therefore, it is necessary to identify the un-modeled dynamics from the experimental results. In this study, we focus on periodic motions such as the welding weaving motion by a powered manipulator and propose an identification method of the structure and parameters of un-modeled dynamic. Based on the periodicity of the motion, the quasi-impulse response of the dynamics is derived from the motion error, and its state-space equation is derived. In particular, since the motion errors include both effects of pre-identified parameter error and un-modeled dynamics, they are separated by determining the dimension of the state-space equation. The effectiveness of the proposed method is evaluated by simulations.

Kinodynamic motion planning is a crucial and challenging topic in robotics for planning motion sequences that satisfy movement constraints, such as autonomous driving and robotic manipulation. Kinodynamic motion planning using templates can quickly explore the search space while satisfying nonholonomic constraints. However, such a heuristic is not effective and robust against diverse environments such as the existence of obstacles. This research introduces a rewiring process in template-based RRT can efficiently find high-quality paths while maintaining a fast sampling speed. The experiment in an artificial environment shows that our proposed approach outperforms the baseline and is able to find better solutions in a shorter time.

In this study, we have discussed the design optimization based on topology optimization for reducing the weight of an interior permanent magnet synchronous motor. The proposed method is based on the genetic algorithm, and objective functions were designed to reduce the weight and keep the average torque. The previous works applied the proposed method to a motor rotor, and in this paper, the proposed method is applied to a motor stator. To use the method for the stator, the coil area has to be constant in all cases because the specification of the motor strongly depends on the coil area on the stator. To realize the constant area of the coil, a novel way to modify the design point of the optimization based on the bisection method is proposed. The validity of the proposed method is shown by applying the proposed method to the practical problem based on the IEEJ D-model.

With the declining birthrate and aging population, the demand for the motorized wheelchair is increasing. A wheelchair that can be instantly moved in all directions by attaching an active-caster had been proposed. This caster enables a lighter and more compact motorized wheelchair than conventional motorized wheelchairs. In this paper, a method for detecting absolute angles using two potentiometers is presented. It leads to smaller gears to be installed and compensates for the dead-zone disadvantages of potentiometers. The absolute angle is determined by using the difference between the output values of multiple potentiometers. However, the influence of the backlash must be considered. Therefore, a gear arrangement with a larger dead zone tolerance is determined. Then, the maximum error in absolute angle caused by backlash is determined.

We investigate the impact of introducing the collision avoidance constraint by the Control Barrier Function (CBF) on search performance of the Particle Swarm Optimization (PSO) algorithm. Swarm robot is superior to individual robots in terms of system flexibility, scalability, and fault tolerance. So, it is expected to effectively perform cooperative tasks, such as searching in a certain area. However, collision avoidance must be considered to ensure system safety and avoid robot failure. Therefore, it is necessary to achieve both search and collision avoidance when performing search on real robots. We discuss the chenge in search performance and search behavior by introducing collision avoidance of CBF.

In this study, we proposed an agent estimation algorithm that reflects sensor inputs for a swarm through local interactions to distinguish the agent with maximum input. Each robot in a swarm is equipped with a phase oscillator. At the proposed algorithm, each robot locally compares its angular velocities which drives each robot’s oscillator and reflects the magnitude of sensory input influenced by attractants with the others and increases its value to catch up with a superior angular velocity. As a result, the angular velocities of all robots converge to a single maximum value. We can identify which robot has the maximum sensory input by focusing on the angular velocity variation and this robot is suitable to the unified leader. This algorithm is mathematically guaranteed by an eigenvalue analysis. We verified the proposed method can distinguish the agent with maximum input by utilizing light as a sensor input in the implementation.

To perform various tasks in the physical world, swarm robots should be able to manipulate objects, such as single object transportation and cooperative operations by multiple individuals. In order to manipulate a variety of objects reliably in various environments, it is necessary to store a lot of experiences of object manipulation by swarm robots and to share the manipulation information. In this study, we describe the verification of actual robots interacting with objects in a real environment and the potential of AR markers, with the aim of finally embedding robot-object interaction by AR markers and QR codes.

At disaster sites, swarm robot systems are required to cooperatively remove obstacles of different number and mass, but an autonomous distributed swarm robot system with scalability, flexibility, and robustness has not been established. In this paper, we examine whether deep reinforcement learning based on Self-Attention can be applied to autonomous distributed swarms of robots to acquire cooperative obstacle removal behavior. Specifically, we verified the robustness, flexibility, and scalability of learning behaviors using the relative coordinates of obstacles and points on the robot’s surface as the robot’s observation information. As a result, it was confirmed that scalability and flexibility can be obtained by the proposed learning method.

In this study, we experimentally verify control methods for bi-connectivity of networks in a multi-robot system using actual robots. It is desirable for the network of a cooperative multi-robot system to be connected since the robots in the system exchange information via the network. To withstand single robot failure, the bi-connectedness of the network is important. The bi-connectedness should be resumed when it is violated. Previous studies proposed control methods for the bi-connectivity and verified them in computer simulations, they did not take into account uncertainties such as noise generated by sensors. Therefore, we experimented with the control method using a real multi-robot system. This paper describes the results of an autonomous decentralized estimation of perturbed algebraic connectivity which is required for the bi-connectivity control and the results of the bi-connectivity control experiments.

We propose a robot identification method for multi-robot systems and report experimental results using the method in this manuscript. Our experimental robot is equipped with full-color LEDs and an omnidirectional camera to detect and identify neighboring robots. Although the color of the LED is used to identify the robots, there is a limit to the number of robot IDs that can be assigned because color recognition depends on ambient light. We propose an approach to alternately flash the LEDs in two colors and identify the robots by recognizing the color combination to increase the number of robot IDs that can be assigned. An ID recognition experiment and a position estimation experiment are conducted to verify the proposed method.

It is difficult to estimate the displacement of a twisted string actuator (TSA), which is a type of linear actuator, using only a mathematical model because the output is nonlinear due to the nature of the viscoelastic resin string. Therefore, in this study, the displacement of the TSA was estimated using a neural network in which the mathematical model of the TSA was introduced. The experimental results showed that the neural network using the mathematical model as cue information improved the accuracy of displacement estimation compared to the results obtained by estimating displacement using only the neural network.

In recent years, there has been a growing demand for insect-scale micro aerial vehicles (MAVs) that can fly in small spaces at disaster sites and agricultural sites, and research on MAVs has been active.In this research, I develop an insect-scale flapping type flying robot using a micro ultrasonic motor with high torque density and simple structure. Therefore, to convert the rotational motion of the motor into a flapping mechanism, I proposed a slider-crank mechanism, and designed, created. The total weight of the parts of the flapping mechanism is less than 30 mg. The frictional torque of the prototype mechanism was 2.17 μNm. The micro ultrasonic motor (3.2 mm × 3.2 mm × 2.0 mm) with a torque of 30 μNm was connected to the mechanism for driving experiments. As the results show that the slider crank mechanism can be driven by a micro ultrasonic motor, indicating its usefulness for a flapping flight robot.

The twisted and coiled polymer-coated optical fiber (TCPOF) is a thermal-driven soft actuator fabricated by twisting and coiling a polymer-coated optical fiber. The TCPOF displacement sensing was validated to be estimated and controlled by using optical fiber embedded in TCPOF, but the displacement sensing considering the load change is remained issue.This paper proposes a TCPOF displacement estimator using the LSTM. Through experiments, we trained the estimator and investigated its performance.

Inverse kinematics is essential for robots to execute tasks, not only for rigid robots but also for robotic surfaces. This paper focuses on a robotic surface that consists of a truncated conical actuator with two air chambers. The actuator can drive the front and back radii independently, and the robot deforms by applying pressure corresponding to the radii obtained by solving inverse kinematics. Since the robotic surface is composed of soft materials, several factors must be investigated to determine the relationship between radius and pressure. In this paper, one air chamber of an actuator is investigated for the pressure of the adjacent air chamber, the presence or absence of a connecting part, and the position of the actuator.

Recently, Model Predictive Path Integral (MPPI) control algorithm has been applied to autonomous navigation tasks. However, using the conventional MPPI algorithm for the autonomous navigation of a quadrotor, there are cases in which a quadrotor cannot stay around a target position or avoid a concave-shaped obstacle. In this paper, we propose two algorithms to solve those problems. The first one is the noise variance control algorithm. This algorithm controls the variance of noise added for sampling depending on the distance between the position of a quadrotor and the target position. The second one is the unapproachable point algorithm. This algorithm creates an unapproachable point if a quadrotor stays in the same position for a while because of unable to avoid an obstacle. An unapproachable point gives a cost depending on the distance from the current position of a quadrotor to keep the quadrotor apart from the unapproachable point. We implemented those two algorithms on simulations and found that the quadrotor can avoid concave-shaped obstacles and stay around the given target position.

One method for obstacle avoidance motion is the potential field method, which determines a robot's movement speed and direction by assigning virtual potential fields to the target location and obstacles. Studies on the potential field method have been conducted extensively, with many defining the attractive and repulsive potential fields as spreading in a circular pattern. However, in our sequentially generated path using a circular potential field, when an obstacle appeared near the robot's side unrelated to the direction of travel, the robot made a large avoidance maneuver, resulting in unstable movement in narrow passages, for example. In this study, we developed a system for sequentially generating paths using an elliptical potential field spreading, with the aim of obstacle avoidance for autonomous mobile robots with preset waypoints. We also compared the obstacle avoidance motion in sequential path generation using a circular potential field and an elliptical potential field.

In this research, we consider a posture control problem of the planer two-link torque unit manipulator under viscous friction condition. The torque unit is a module which consists of an electric motor and a reaction wheel. The friction arising from the presence of a viscous environment cannot be ignored in the real system. In this report, we show by numerical simulation that the rotation angle of the second link can be controlled to 0 by using a periodic input. Next, as a verification experiment, we confirm that the first link and the second link perform periodic motion by giving a periodic input to the torque unit. Then we show that the rotatation angle of the second link can be controlled to 0 by a verification experiment.

This paper presents vibration-free positioning control of a double pendulum driven by a linear actuator, which is applicable to vibration suppression of a slung load carried by unmanned aerial vehicle (UAV). Since the system is not differentially flat, a linear flat output for its linear approximation is used to generate the vibration-free trajectory for the linear actuator. The flat output is firstly characterized based on the Brunovsky canonical form of the control-affine system, then a simple formula to determine the linear flat output using system matrices is derived. The approximate flatness-based controller suppresses residual vibration both in numerical simulations and experiments.

There is a demand for robots that can work in place of humans with the problem of labor shortages in Japan’s industrial sites. However, motion teaching to the robot is very time consuming with conventional methods and a heavy burden on the instructor. Therefore, we aim to make the motion path generation automatic in order to reduce the burden on the instructor. The path generation generates a path that does not contact any obstacles. In our proposed method, the instructor only needs to set the transit point and parameters, and the path is automatically corrected based on the force information obtained at the time of contact. The parameter adjustment process is difficult, so it will be necessary to consider making it automatic in the future.

This article focuses on the driving mechanism of a miniature soft actuator for endoscopic surgery. The actuator is designed to perform deploying/retraction and stiffening/softening behavior with hydraulic control. A temperature-responsive shape memory polymer embedded as a backbone to adjust the stiffness and keep the shape of the actuator. We first introduce the design and control strategy, and then show the finite element method results of its retraction behavior. Next, we illustrate the fabrication process of actuator prototypes, and report the deploying/retracting demonstration experiment. Our work primarily proves the actuation ability of the actuator, and reveals the direction of future improvement.

This paper reports on the detection of cells using a micropipette integrated with a capacitive microsensor fabricated by a hybrid fabrication process that combines MEMS fabrication technology and 3D printing. The micropipette integrated with a capacitive microsensor was fabricated by 3D printing directly onto a MEMS device. First, the electrodes of capacitive microsensor were fabricated using a conventional MEMS process. Next, the pipette structure was 3D printed directly on the substrate using projection micro stereolithography with UV curable resin. Then, the sensor substrate was cut into pipette shapes. We demonstrated the use of the fabricated device to detect cell.

Recently micro-fluidic devices have attracted much attention especially for applications in the biotechnology and chemical engineering. The advantage of microfluidic devices is that fewer biological materials and chemicals are needed for experiments. To achieve high functions in these devices, it is indispensable to equip robust and compact micropump elements. In this work, we produced the propeller used for micro-pump driven by nanosecond laser-induced bubbling. When intense laser pulses are focused on a liquid in the microfluidic device, a jet flow is induced with shockwave and stress generation, which is a source of the bubbling. The focused laser pulse was utilized as an ignitor to rotate propeller in the device i.e., this micro-pump is not required to equip ignitor in the device.

In this study, a cell homogeneous dispersion structure with low cell loss in microfluidic devices was developed. Microfluidic devices can exert great power in experiments using biological cells, but the small scale of the channels prevents the uniform introduction of cells, reducing the efficiency and reproducibility of the experiments. The use of micropillar arrays (MPA), in which the spacing between the pillars is smaller than the diameter of the cells, generates random flow as cells are repeatedly embolised and passed between the pillars, resulting in uniform dispersion, but the low cell passage rate makes it inapplicable to valuable cells. In this study, a balloon was used to switch the flow channel width in MPAs with sufficient pillar spacing for the cell size to achieve both uniform dispersion of cells and an increase in cell passage rate.