Published papers

In orthopedic surgery, making an incision into the spine involves a risk of injury to the spinal cord. In addition, surgeons must determine penetration into the bone using only their haptic senses. This imposes a heavy burden on the surgeon. In this study, an orthopedic haptic drill with a penetration detection scheme based on viscosity estimation is proposed. This drill detects penetration by monitoring changes in the position and force of the linear motor in the drill. The threshold values are automatically optimized according to the estimated viscosity of the object being cut. Therefore, the proposed drill does not require prior setup of the parameters in accordance with the object being cut. The utility of the proposed drill is verified experimentally.

Robotic grinding is widely used for shaping workpieces in manufacturing, but it remains difficult to automate this process efficiently. In particular, efficiently grinding workpieces with different shapes and density-dependent removal resistance is challenging because removal resistance varies with local contact conditions. Moreover, it is difficult to achieve accurate estimation of removal resistance and analytical modeling of shape transition, and learning-based approaches often require large amounts of training data to cover diverse processing conditions. To address these challenges, we decompose robotic grinding into two components: removal-shape planning and contact-force adaptation. Based on this formulation, we propose DecompGrind, a framework that combines Global Cutting-Surface Planning (GCSP) and Local Contact-Force Adaptation (LCFA). GCSP determines removal shapes through geometric analysis of the current and target shapes without learning, while LCFA learns a contact-force adaptation policy using bilateral control-based imitation learning during the grinding of each removal shape. This decomposition restricts learning to local contact-force adaptation, allowing the policy to be learned from a small number of demonstrations, while handling global shape transition geometrically. Experiments using a robotic grinding system on 3D-printed workpieces with different shapes and densities under fixed belt conditions demonstrate reduced grinding time while maintaining safe levels of contact force.

We propose a sensorless remote catheter control system that delivers real-time haptic feedback. Bilateral control combined with a reaction force observer estimates contact forces and transmits them to the operator in real time, while frequency-shaped feedback improves the perceived sensitivity and reduces fatigue. Using three catheter types and three operators, we evaluated position-force synchronization, loop delay, and vessel-wall load relative to manual manipulation. The average position error was approximately 2.14× 10-5m, and the force error was 0.493 N. Cross-correlation (≤ 1Hz) showed no measurable lag in the low-frequency synchronization analysis within the 1 ms resolution. For force equivalence, the prespecified primary analysis employed a two one-sided tests (TOST) procedure based on estimates obtained from a linear mixed-effects (LME) model, with the smallest effect size set to ±10% of the manual mean; equivalence was supported (b = -0.029, 90% confidence interval [-0.119, 0.061]), whereas a stricter ±5% sensitivity analysis was not supported. All operators completed the tasks after minimal system familiarization. These findings indicate that the system replicates manual precision in a benchtop phantom setting while enabling real-time haptic feedback and maintaining vessel-wall loading comparable to that of manual manipulation. Our sensorless haptic technology represents a promising step toward future clinical translation of remote catheter systems, with applications in safer, standardized, and more intuitive endovascular procedures.

Master-slave teleoperatìon robots are useful for tasks that require human judgments and skills. Haptic sensation can be transmitted between a master and a slave by implementing bilateral control. Besides, visual information of the slave side that cannot be directly seen from the master side is presented by devices such as cameras, endoscopes, and displays. Therefore, advanced support of the teleoperatìon robots for operators can be expected by utilizing extended visual and haptic information. In this paper, a master-slave teleoperation system with ability to extend information transmitted to operators is proposed. The teleoperation system consists of one degree of freedom master-slave system implementing bilateral control based on robust acceleration control and an image processing system. To support operators, visual and haptic augmented reality is generated by image processing, micro-macro bilateral control, and compliance control. In addition, a design method of position and force scaling for micro-macro bilateral control according to magnification ratio of image is presented. The utility of the proposed bilateral teleoperation system is verified by experiments.

Surgical robots have been studied and developed to assist surgeons. Master-slave flexible forceps robots that are components of a flexible endoscopic surgical robot are useful to realize novel minimally invasive surgical procedures that are difficult for conventional rigid medical devices. Although driving force for the tip portion of a flexible forceps robot is mechanically transmitted by wires through the flexible portion, friction and backlash changed by the posture of the flexible portion make it difficult to realize precise end effector control. Furthermore, transmission of haptic sensation is necessary to realize safer operation. In this study, a multi degrees-of-freedom (DoF) haptic forceps robot with three fingers is presented as an end effector of a flexible forceps robot. Transmission of driving force in flexible structure that causes performance deterioration is eliminated by arranging motors near the end effector. To transmit haptic sensation to operators, bilateral control is implemented to a master-slave system composed of the haptic forceps robot and a multi DoF input device. Additionally, control on modal space is implemented to verify finger dexterity. The utility of the haptic forceps robot is experimentally validated.

This article analyses the robust stability and performance of the disturbance observer (DOb) based digital motion control systems in the discrete-time domain. It is shown that the phase margin and the robustness of the digital motion controller can be directly adjusted by tuning the nominal plant model and the bandwidth of the observer. However, they have the upper and lower bounds due to robust stability and performance constraints as well as the noise sensitivity. The constraints on the design parameters of the DOb change when the digital motion controller is synthesized by measuring different states of a servosystem. For example, the bandwidth of the DOb is limited by the noise sensitivity and waterbed effect when the velocity and position measurements are employed in the digital robust motion controller synthesis. The robustness constraint due to the waterbed effect is removed when the DOb is implemented by acceleration measurement. The design constraints on the nominal plant model and the bandwidth of the observer are analytically derived by employing the generalized Bode integral theorem in discrete time. The proposed design constraints allow one to systematically synthesize a high-performance DOb-based digital robust motion controller. The experimental results are given to verify the proposed analysis and synthesis methods.

The force/tactile sensation (hereafter referred to as f/t sensation) has not been well defined for a long time. In the applications of real haptics, it is necessary to quantitatively evaluate the quality of the f/t sensation. The paper proposes the definition of f/t sensation as the stimulus signal. Accordingly, the instantaneous value of the stimulus signal of the f/t sensation represents the simultaneous ratio of the value of reaction force and the value of velocity. If the contact target is modeled as an inertia-damper-spring model, the stimulus signal of the f/t sensation is expressed as the equivalent circuit of the sum of the impedance (or admittance) and the voltage source (or the current source). The definition of the stimulus signal of the f/t sensation in frequency area clarifies the physical meaning. The paper presents the experimental results that confirm the viability of this novel definition on the stimulus signal of the f/t sensation.

Robots are being developed to perform tasks in homes and factories autonomously. Several studies have examined motion generation based on haptic information, and some studies consider the environment as physical property information. However, there is a trade-off between the accuracy and the time required for the physical property estimation. Therefore, in this study, we propose a method for estimating physical properties based on training the relationship between the two estimation models. The first is a fast sequential estimation model, and the second is a highly accurate posterior estimation model. Training the relationship between the two models makes highly accurate sequential property estimation possible. Validation results showed improved accuracy of property estimation for learning samples and some untrained samples.

Teleoperation with hydraulic actuator is useful for human action augmentation. However, disturbances in hydraulic actuators are complex and accurate estimation of external forces is difficult. In this paper, a reaction force observer and machine learning are combined to achieve high accuracy sensorless force estimation in hydraulic actuator. Furthermore, this method is applied to a bilateral control system to improve its performance. While there are many machine learning methods, this paper uses a Long Short-Term Memory network, a type of recurrent neural network that excels at inferring time series data, to accurately infer the hysteresis characteristics of disturbances in hydraulic actuator. Furthermore, 4ch bilateral control based on oblique coordinate control is used to realize teleoperation. In the experiment, a friction model-based compensation method and a machine learning-based compensation method are applied to bilateral control, and the performance of each method is evaluated.

This paper presents a novel inductive position sensor designed for motion control applications in harsh environments. Traditional angle sensing technologies such as optical and magnetic encoders often suffer from performance degradation under extreme conditions including high temperature, humidity, and radiation. The proposed sensor operates based on electromagnetic inductance variation caused by air-gap modulation, offering robust and contactless angle measurement. Its integration into position and velocity control loops of a rotary motor demonstrates the sensor's capability to provide reliable feedback for closed-loop control. Experimental validation confirms stable and accurate motor performance, supporting the sensor's applicability in industrial and robotic systems operating under challenging environmental conditions.

This paper presents a novel angle measurement sensor designed for operation in harsh environments. Typically, angle measurement is performed using encoders that are directly attached to motors, actuators, or both, depending on the application. These encoders can be either incremental or absolute and may utilize various sensing principles, such as optical, magnetic, or resistive methods. However, most conventional sensors exhibit significant limitations when deployed in extreme conditions, such as radioactive environments, high temperatures, or high moisture levels. In such scenarios, robotic systems often experience failures due to sensor malfunctions, making the development of robust angle measurement solutions essential for ensuring fail-safe operation. This paper introduces a novel angle measurement sensor based on the variation of electromagnetic inductance, specifically designed for use in harsh environments. Experimental results obtained from a laboratory setup validate the effectiveness of the proposed method.

The current methods to generate robot actions for automation in significantly different environments have limitations. This paper proposes a new method that matches the impedance of two prerecorded action data with the current environmental impedance to generate highly adaptable actions. This method recalculates the command values for the position and force based on the current impedance to improve reproducibility in different environments. Experiments conducted under conditions of extreme action impedance, such as position and force control, confirmed the superiority of the proposed method over existing motion reproduction system. The advantages of this method include the use of only two sets of motion data, significantly reducing the burden of data acquisition compared with machine-learning-based methods, and eliminating concerns about stability by using existing stable control systems. This study contributes to improving the environmental adaptability of robots while simplifying the action generation method.

The risk of spinal cord injury when cutting into the spine is a particular concern in orthopedic surgery. Surgeons must rely entirely on their haptic senses to determine bone penetration. Therefore, surgeons incur heavy burdens. Some studies have introduced penetration-detection systems using torque information. However, these method requires obtaining the characteristics of the bone before the actual cutting. Our previous study introduced an orthopedic haptic drill equipped with a penetration-detection system. This drill detects penetration by monitoring changes in the force of a linear motor. In this study, the advantage of using force over torque for penetration detection was verified through theoretical modeling and experiments. The theoretical model indicates that the thrust force reacts more sensitively than the torque at penetration. The detection methods using the force-based and torque-based methods were compared through experiments. The experimental results verified that the force-based detection method has advantages in terms of detection accuracy and detection time. The results demonstrated the efficacy of the proposed method, highlighting its potential utility in improving the accuracy and safety of orthopedic procedures.

In orthopedic surgery, robotic technologies to assist surgeons operating surgical drills to cut bones are being de-veloped, since there is a risk to injure delicate tissues around bones. Penetration detection is useful for surgical drills to enhance safety. However, there is still a risk to damage delicate tissues, even if penetration is detected to stop drills. In this study, a teleoperated haptic drill system implementing a safety enhancement system which extends haptic sensation according to the end effector position is presented. The presented method is composed of the object detection phase and the operator assist phase. In the object detection phase, the position and posture of an object are acquired by the drill system which has the shape model of the object. In the operator assist phase, haptic sensation is extended based on the acquired object information to prevent penetration of the object, while the state of the controller of the drill system is visualized by CG images created by a cutting simulator. The feasibility of the method is confirmed by cutting a pseudo bone.

Understanding human movements is essential for teaching skills to novices and enabling robots to learn. Traditional movement analysis often relies on visual data, but incorporating force/tactile information is vital for object interactions. The spring-mass-damper model, used to represent force/tactile data, assumes continuous contact and consistent material properties, which are limitations when analyzing grinding actions that frequently alternate between contact and non-contact. This study proposes analyzing grinding by quantifying force/tactile sensations as the ratio of the root mean square (RMS) values of force to speed, using absolute impedance to measure sensations irrespective of contact. Findings indicated that operators typically initiate a pressing motion several seconds after detecting contact, and their reactions to samples of varying hardness differ significantly. These observations suggest a detailed breakdown of human grinding skills is feasible, highlighting the potential of force/tactile information in enhancing skill transfer to beginners and robots.

The stiffness of human cancers may be correlated with their pathology, and can be used as a biomarker for diagnosis, malignancy prediction, molecular expression, and postoperative complications. Neurosurgeons perform tumor resection based on tactile sensations. However, it takes years of surgical experience to appropriately distinguish brain tumors from surrounding parenchymal tissue. Haptics is a technology related to the touch sensation. Haptic technology can amplify, transmit, record, and reproduce real sensations, and the physical properties (e.g., stiffness) of an object can be quantified. In the present study, glioblastoma (SF126-firefly luciferase-mCherry [FmC], U87-FmC, U251-FmC) and malignant meningioma (IOMM-Lee-FmC, HKBMM-FmC) cell lines were transplanted into nude mice, and the stiffness of tumors and normal brain tissues were measured using our newly developed surgical forceps equipped with haptic technology. We found that all five brain tumor tissues were stiffer than normal brain tissue (p < 0.001), and that brain tumor pathology (three types of glioblastomas, two types of malignant meningioma) was significantly stiffer than normal brain tissue (p < 0.001 for all). Our findings suggest that tissue stiffness may be a useful marker to distinguish brain tumors from surrounding parenchymal tissue during microsurgery, and that haptic forceps may help neurosurgeons to sense minute changes in tissue stiffness.

Bone cutting tasks in spinal surgery require advanced skills to operate a handheld drill device. Since there is a risk to injure the spinal cord, both of support for surgeons and enhancement of surgeon's skills are important for safer surgery. Therefore, orthopedic surgical robots and devices have been studied and developed. In this study, a handheld haptic drill simulator using a visual servoing system is developed. A stereo camera is used to acquire the position and posture of a handheld drill device used as a haptic interface. Voxel representation is applied to virtual objects in environment simulation. Axial cutting force is fed back to the operator via the haptic interface, while virtual environment images are presented by a two-dimensional display. A small virtual object composed of voxels is cut by the drill device in the experiment. The feasibility of the handheld haptic drill simulator and the applicability of the handheld haptic drill devices are confirmed by the experimental results.

Electrohydrodynamics (EHD) is a thrust force produced when high voltage is applied to two electrodes, and actuators using EHD are expected to reduce equipment size and the effects of friction. However, the EHD effect of multiple anode electrodes has yet to be verified and should be considered when fabricating a multi-degree-of-freedom actuator using EHD. This paper examines the relationship between the applied voltage and the generated EHD to verify the effect of multiple anode electrodes on the EHD. Experiments were conducted using a device with two anode electrodes at right angles and another with two parallel electrodes. The results showed that the two anode electrodes placed at right angles interfered with each other, and the generated forces were not superposed. The direction of the generated force was approximately 45 deg. When the electrodes were placed horizontally, it was confirmed that the generated forces were superposed, indicating that the arrangement, not the number of anode electrodes, was the cause of the interference.

Real haptics is a technology that reproduces the sense of force and touch by transmitting contact information with real objects by converting human movements and the feel of the objects into data. In recent years, real haptics technology has been installed in several surgical devices. A custom-made surgical drill was used to drill into the posterior lamina to verify the time required for penetration detection and the distance the drill advanced after penetration. A surgeon operated with the drill and the same aspects were measured and verified. All experiments were performed on female miniature pigs at 9 months of age with a mean body weight of 23.6 kg (range 9–10 months and 22.5–25.8 kg, n = 12). There were statistically significant differences in the average reaction time and the distance travelled after penetration between a handheld drill and the drill with the penetration detection function (p < 0.001). The reaction time to detect penetration and the distance after penetration were both significantly improved when compared with those of the handheld surgical drill without the penetration detection function, with mean differences of 0.049 ± 0.019 s [95% CI 0.012, 0.086 s] and 2.511 ± 0.537 mm [95% CI 1.505, 3.516 mm]. In this study, we successfully conducted a performance evaluation test of a custom-made haptic interface surgical drill. A prototype high-speed drill with a haptic interface accurately detected the penetration of the porcine posterior lamina.