One of the goals in the field of mobile robotics is the development of personal service robots for the elderly which behave in populated environments. In this paper, we describe a security robot system and ongoing research results that minimize the risk of the elderly and the infirm to access an area to enter restricted areas with high potential for falls, such as stairs, steps, and wet floors. The proposed robot system surveys a potential falling area with an equipped laser scanner sensor. When it detects walking in elderly or infirm patients who in restricted areas, the robot calculates the velocity vector, plans its own path to forestall the patient in order to prevent them from heading to the restricted area and starts to move along the estimated trajectory. The walking human is assumed to be a point-object and projected onto a scanning plane to form a geometrical constraint equation that provides position data of the human based on the kinematics of the mobile robot. While moving, the robot continues these processes in order to adapt to the changing situation. After arriving at an opposite position to the human's walking direction, the robot advises them to change course. The simulation and experimental results of estimating and tracking of the human in the wrong direction with the mobile robot are presented.

In this paper, we designed and tested an ankle joint mechanism for a gait rehabilitation robot. Gait rehabilitation programs are designed to improve the natural leg motion of patients who have lost their walking capabilities by accident or disease. Strengthening the muscles of the lower-limbs and stimulation of the nervous system corresponding to walking helps patients to walk again using gait assistive devices. It is an obvious requirement that the rehabilitation system's motion should be similar to and as natural as the normal gait. However, the system being used for gait rehabilitation does not pay much attention to ankle joints, which play an important role in correct walking as the motion of the ankle should reflect the movement of the center of gravity (COG) of the body. Consequently, we have designed an ankle mechanism that ensures the safety of the patient as well as efficient gait training. Also, even patients with low leg muscle strength are able to operate the ankle joint due to the direct-drive mechanism without a reducer. This safety feature prevents any possible adverse load on the human ankle. The additional degree of freedom for the roll motion achieves a gait pattern which is similar to the normal gait and with a greater degree of comfort.

For the self-directed rehabilitation of upper extremity hemiplegia patients, in this paper we propose an interface method capable of doing bilateral exercises in rehabilitation robotics. This is a method for estimating information of movements from the unaffected-side, and projects it to the affected-side in order. That the affected-side is followed the movements of the unaffected-side. For estimation of the unaffected-side movements information, gyro sensor data and acceleration sensor data were fused. In order to improve the measurement error in data fusion, a HDR filter and a complementary filter were applied. Estimated motion information is derived the one side of the drive input of rehabilitation robot. In order to validate the proposed method, experimental equipment is designed to be similar to the body's joints. The verification was performed by comparing the estimation angle data from inertial sensors and the encoder data which were attached to the mechanism.

This paper proposes a gait phase classifier using a Recurrent Neural Network (RNN). Walking is a type of dynamic system, and as such it seems that the classifier made by using a general feed forward neural network structure is not appropriate. It is known that an RNN is suitable to model a dynamic system. Because the proposed RNN is simple, we use a back propagation algorithm to train the weights of the network. The input data of the RNN is the lower body's joint angles and angular velocities which are acquired by using the lower limb exoskeleton robot, ROBIN-H1. The classifier categorizes a gait cycle as two phases, swing and stance. In the experiment for performance verification, we compared the proposed method and general feed forward neural network based method and showed that the proposed method is superior.

Early wearable robotic devices were big, powerful and manipulator-like. Recently, various applications of wearable robotics have shown a greater demand for lower weight and compliancy. One approach to achieve these objectives is the use of novel actuators such as twisted string actuators. These actuator are very light, quiet, mechanically simple and compliant. Therefore, they can drastically decrease the weight and size of robotic systems such as exoskeletons. However, one drawback of this actuator is its nonlinear transmission ratio, which is established as a ratio between the angle of twisting of the strings and their resulting contraction. In this paper, we propose a transmission mechanism with rotational motion as the output incorporating a twisted string actuator (TSA). The designed mechanism allows the linearization of the relationships between the input and output displacements and forces of a TSA. The proposed design has been validated theoretically and through a set of computer simulations. A detailed analysis of the performance of the proposed mechanism is presented in this paper along with a design guideline.

Hand function is one of the essential functions required to perform the activities of daily living, and wearable robots that assist or recover hand functions have been consistently developed. Previously, wearable robots commonly employed conventional robotic technology such as linkage which consists of rigid links and pin joints. Recently, as the interest in soft robotics has increased, many attempts to develop a wearable robot with a soft structure have been made and are in progress in order to reduce size and weight. This paper presents the concept of a soft wearable robot composed of a soft structure by comparing it with conventional wearable robots. After that, currently developed soft wearable robots and related issues are introduced.

This paper describes an intelligent ankle assistive robot which provides assistive power to reduce ankle torque based on an analysis of ankle motion and muscle patterns during walking on level and sloped floors. The developed robot can assist ankle muscle power by driving an electric geared motor at the exact timing through the use of an accelerometer that detects gait phase and period, and a potentiometer to measure floor slope angle. A simple muscle assistive link mechanism is proposed to convert the motor torque into the foot assistive force. In particular, this mechanism doesn't restrain the wearer's ankle joint; hence, there is no danger of injury if the motor malfunctions. During walking, the link mechanism pushes down the top of the foot to assist the ankle torque, and it can also lift the foot by inversely driving the linkage, so this robot is useful for foot drop patients. The developed robot and control algorithm are experimentally verified through walking experiments and EMG (Electromyography) measurements.

In this paper, an ARS-EKF based motion counting algorithm for repetitive exercises such as calisthenics is proposed using a smartwatch. Raw sensor signals from accelerometers and gyroscopes are widely used for conventional smartwatch counting algorithms based on pattern recognition. However, generated features from raw data are not intuitive to reflect the movement of motions. The proposed motion counter algorithm is composed of navigation based feature generation and counting with error correction. The candidate features for each activity are velocity and attitude calculated through an ARS-EKF algorithm. In order to select those features which reveal the characteristics of each motion, an exercise frame from the initial sensor frame is introduced. Counting processes are basically based on the zero crossing method, and misdetected counts are eliminated via simple classification algorithms considering the frequency of the counted motions. Experimental results show that the proposed algorithm efficiently and accurately counts the number of exercises.

The purpose of this research is to develop a gait assistive device to enhance the gait stability and training efficiency of stroke patients. The configuration of this device is mainly composed of a motored wheel and a single cane whose lower end is attached to a motored wheel frame. A patient can feel haptic information from continuous ground contact from the wheel while walking through the grip handle. In addition, the wheeled cane can avoid using excessive use of the patient's upper limb for weight support and motivate the patient to use a paralyzed lower limb more actively. Moreover, the proposed device can provide intuitive and safe user interaction by integrating a force sensor and a tilt sensor equipped to the cane frame, and a switch sensor at the cane's handle. The admittance control has been implemented for the patient to change the walking speed intuitively by using the interaction forces at the handle. A hemi-paretic stroke patient participated in the walking assistive experiments as a pilot study to verify the effectiveness of the proposed haptic cane system. The results showed that the patient could improve walking speed and muscle activations during walking with a constant speed mode of the haptic cane. Moreover, the patient could maintain the preferred walking speeds and gait stability regardless of the magnitude of resistance forces with the admittance control mode of the haptic cane. The proposed robotic gait assistive device with a simple and intuitive mechanism can provide efficient gait training modes to stroke patients with high possibilities of widespread utilizations.

This paper presents a design method of a static output feedback controller for continuous T-S fuzzy systems via parallel distributed compensation (PDC). The existence condition of a set of static output feedback gains is represented in terms of linear matrix inequalities (LMIs). The sufficient condition presented here does not need any transformation matrices and equality constraints and is less conservative than the previous results seen in [20].

Supervised machine learning has become popular in discovering context descriptions from sensor data. However, collecting a large amount of labeled training data in order to guarantee good performance requires a great deal of expense and time. For this reason, semi-supervised learning has recently been developed due to its superior performance despite using only a small number of labeled data. In the existing semi-supervised learning algorithms, unlabeled data are used to build a graph Laplacian in order to represent an intrinsic data geometry. In this paper, we represent the unlabeled data as the spatial-temporal dataset by considering smoothly moving objects over time and space. The developed algorithm is evaluated for position estimation of a smartphone-based robot. In comparison with other state-of-art semi-supervised learning, our algorithm performs more accurate location estimates.

This paper proposes an auto-parking controller for omnidirectional mobile robots. The controller uses the multi-sensor system including ultrasonic sensor and camera. The several ultrasonic sensors of robot detect the distance between robot and each wall of the parking lot. The camera detects the global position of robot by capturing the image of artificial landmarks. To improve the accuracy of position estimation, we applied the extended Kalman filter with adaptive fuzzy controller. Also we developed the fuzzy control system to reduce the settling time of parking. The experimental results are presented to verify the usefulness of the proposed controller.

This paper proposes an efficient detection of absolute position of a robot joint using two incremental encoders. We considers a robot joint comprising a motor, a reducer, two encoders, and a motor drive. An incremental(first) encoder provides motor's rotor position or input position of reducer while another incremental(second) encoder does output position of the reducer. A table is made where the relationship between the first and the second encoder counts is recorded. The key point is placed where the table is constructed: when a pulse occurs in the second encoder, there exists a corresponding unique count value of the first encoder. The absolute position is detected using the table by searching the second encoder position corresponding to the first encoder count value when a pulse occurs in the second encoder. The proposed method needs a small rotation, as just one second encoder's pulse angle, for the initial absolute position detection.

Flexible tactile sensors can provide valuable feedback to intelligent robots regarding the environment around them. This is especially important when robots such as, service robots share a workspace with humans. This paper presents a contact force measurement algorithm of a flexible tactile sensor. This sensor is manufactured by a direct-writing technique, which is one 3D printing method, using multi-walled carbon nano-tubes. An analog signal processing circuit has been designed and implemented to reduce noise contained in the sensor output. In addition, a digital version of the Butterworth filter was implemented by software running on a microcontroller. Through various experiments, characteristics of the sensor system have been identified. Based on three traits, an algorithm to detect the contact and measure the contact force has been developed. The entire system showed a promising prospect to detect the contact over a large and curved area.

In this paper, we described the Cramer-Rao Lower Bound (CLRB) performances of Time Difference of Arrival (TDOA) and Frequency Difference of Arrival (FDOA) methods when there are multiple emitters. The TDOA and FDOA values between two receivers can be simultaneously estimated by using the so-called Complex Ambiguity Function (CAF). In the case of multiple emitters, there exist Inter Symbol Interferences (ISIs) in the measurement data. Therefore, it is required to reduce the effect of ISI and provide a performance evaluation method of TDOA and FDOA estimations. In order to eliminate the ISIs, using of a filter bank before calculating CAF is proposed when the carrier frequencies of the emitters are different to one another. Angle of Arrival (AOA) or Received Signal Strength (RSS) methods before calculating CAF were proposed to reduce the ISIs when the carrier frequencies are the same. In order to evaluate the CRLB of TDOA and FDOA estimations, we employed the conditional probability distribution method and described the numerical comparison results.