Rong-Jyue Wang

Image Recognition and Force Measurement Application in the Humanoid Robot Imitation

This paper studies and implements motion imitation interaction between a humanoid robot and a human. To achieve the implementation, three main processes are required. The first is the human motion data acquisition; the second is the motion data modification; and the last is the ankle angle adjustment on the supporting foot (or feet) of the humanoid robot. In the human motion data acquisition, we use a webcam to recognize the 13 marks that are pasted on the humans body and then calculate and record the relative positions of the marks for each motion into the motion database. In the motion data modification, the recorded motion data are modified by computer simulation to guarantee that the zero moment point of the robot is inside the stable region. Finally, based on the force sensor measurement on the soles of the robot, the ankle angles of the supporting foot (or feet) are adjusted such that the humanoid robot is capable of imitating the human motion with a balance status in real time. The experimental results demonstrate that the humanoid robot successfully imitates a series of human gymnastic motions.

A Course in Simulation and Demonstration of Humanoid Robot Motion

An introductory course for humanoid robot motion realization for undergraduate and graduate students is presented in this study. The basic operations of AX-12 motors and the mechanics combination of a 16 degrees-of-freedom (DOF) humanoid robot are presented first. The main concepts of multilink systems, zero moment point (ZMP), and feedback control are then introduced such that the students can understand the relationship between the ZMP position and the stability of the robot. Finally, the students can simulate the desired motion trajectory and realize the motion practically on the real humanoid robot. Taking this course, the students can not only learn robotic theories and control techniques for humanoid robot motion, but can also enhance their experience in hands-on experiments in executing the motion of a humanoid robot. The proposed educational strategy will enable students to progress easily to more advanced work on robot design and control in their future study or careers.

The multiple-function intelligent robotic arms

This paper details the design, production, and programming methodology of a multiple-function robotic arm. All of the hardware and software of this robotic arm were designed and produced by the authors. This robotic arm won the championship of the first competition of HIWIN intelligent robotic arms on Aug. 22, 2008 in Taiwan. The main design goal of this robotic arm was to present the following functions: fancy dancing, weightlifting, Chinese calligraphy, and color classification. Another design goal was to minimize cost and maximize performance. On the other hand, a set of the robotic arms are also applied to show martial arts and play the rock-scissors-paper game with modifying and mimetic hands. The characteristics of the robotic arms include: 1. The shoulder of robotic arm includes a pair of motor structures to enhance the ability to life weight. 2. The stability and accurateness of the robotic arm are optimized for the requirement of high performance throughout the whole structural design. 3. In order to increase the moving ability of the robot, the robotic arm was designed with a four-wheeled transmission structure and track. 4. Two mimetic robotic arms work in concert to present the fancy shows. 5. Five kinds of machine hands were designed to meet the requirements of the six appointed functions.

Human Posture Recognition Based on Images Captured by the Kinect Sensor

In this paper we combine several image processing techniques with the depth images captured by a Kinect sensor to successfully recognize the five distinct human postures of sitting, standing, stooping, kneeling, and lying. The proposed recognition procedure first uses background subtraction on the depth image to extract a silhouette contour of a human. Then, a horizontal projection of the silhouette contour is employed to ascertain whether or not the human is kneeling. If the figure is not kneeling, the star skeleton technique is applied to the silhouette contour to obtain its feature points. We can then use the feature points together with the centre of gravity to calculate the feature vectors and depth values of the body. Next, we input the feature vectors and the depth values into a pre-trained LVQ (learning vector quantization) neural network; the outputs of this will determine the postures of sitting (or standing), stooping, and lying. Lastly, if an output indicates sitting or standing, one further, similar feature identification technique is needed to confirm this output. Based on the results of many experiments, using the proposed method, the rate of successful recognition is higher than 97% in the test data, even though the subjects of the experiments may not have been facing the Kinect sensor and may have had different statures. The proposed method can be called a “hybrid recognition method”, as many techniques are combined in order to achieve a very high recognition rate paired with a very short processing time.

Implementation of an Object-Grasping Robot Arm Using Stereo Vision Measurement and Fuzzy Control

In this paper, a method using a stereo vision device and fuzzy control to guide a robot arm to grasp a target object is proposed. The robot arm has five degrees of freedom including a gripper and four joints. The stereo vision device located beside the arm captures images of the target and the gripper. Image processing techniques such as color space transformation, morphologic operation, and 3-D position measurement are used to identify the target object and the gripper from the captured images and estimate their relative positions. Based on the estimated positions of the gripper and the target, the gripper can approach and grasp the target using inverse kinematics. However, since the robot arm’s accuracy of movement may be affected by gearbox backlash or hardware uncertainty, the gripper might not approach the desired position with precision using only inverse kinematics. Therefore, a fuzzy compensation method is added to correct any position errors between the gripper and target such that the gripper can grasp the target. Using the proposed method, the stereo vision device can not only locate the target object but also trace the position of the robot arm until the target object is grasped. Finally, some experiments are conducted to demonstrate successful implementation of the proposed method on the robot arm control.

Vision based 3-D position control for a robot arm

This paper presents a 3-D position control for a robot arm. The system contains a fabricated robot arm, a pair of charge-coupled device (CCD) cameras, and a computer. The inverse kinematics (IK) concept is utilized to manipulate the robot arm. The two-CCD vision geometry is utilized to measure the practical 3-D position of the robot arms tip. Furthermore, a fuzzy position error compensator is added to adjust the target position for the IK technique such that the position accuracy can be guaranteed. The experimental results demonstrate that the robot arm can position its tip at the desired position accurately.

A real time terrain recognition method for mobile robot moving

This study proposes a real time terrain recognition method for a mobile robot moving on uneven terrain. A RGB-D sensor called XtionPro is mounted on the top layer of robot to capture terrain depth image. To estimate the captured terrain depth image, a fast terrain estimation method based on virtual depth image technique and image subtraction is proposed. The proposed method provides an accurate elevation map of the front terrain. Finally, according the leading edge extracted from the elevation map, the robot can recognize that the front terrain is travelable or un-travelable.

Fuzzy control-based real-time robust balance for a humanoid robot

Abstract This paper studies and implements a real-time robust balance control for a humanoid robot under three environment disturbances which are an external thrust, an inclinable platform, and a see-saw. More precisely to say, the robot with robust control can resist an external thrust, stand on a two-axis inclinable platform, or walk on a see-saw successfully. The main feature of the robot is that it has a waist joint which has three degrees of freedom. With the aids of the proposed fuzzy controllers, the robot can change the posture of the body nimbly by adjusting the waist joint and two ankle joints to strengthen the stabilization capacity. The sensory system of the robot includes eight force sensors and one inertial measurement unit sensor in order to measure the center of pressure and the slant angle of the robot’s body. According to the measured data from the sensors and by imitating human reflex actions, the proposed fuzzy controllers perform real-time balance control for the robot under three environment disturbances. According to the experiment results, the stability of the robot is increased at least 32.2 and 61.7% under the first two environment disturbances, respectively. In addition, the robot walking on a see-saw has a success rate of about 95%.

Power Management and Charging Method for the Battery-Set of the Intelligent Servant Robot

This paper studies and implements “high efficiency power management system and intelligent battery-set charger system” for “the Intelligent Servant Robot”. This power management system provides high quality and efficiency electric power supply for all subsystems of the servant robot and accurately estimates the residual capacity of battery set system of the servant robot. This servant robot will be charged by the intelligent battery-set charger when the residual capacity of battery sets is insufficient. This servant robot can take the elevator, follow the user, carry articles, and provide the service of campus security patrolling and monitoring. LiFePO4 batteries are chosen as main power supply of the servant robot. The LiFePO4 battery-set system is divided to two subsystems which are the power battery-set (36V/20AH) for motors and the system battery-set (48V/10AH) for PC and MCUs, respectively. High efficiency power management system can measure the basic information (voltage, temperature, and current), provide protection, and give alarm for battery-set system. For the power converter module design of the intelligent battery-set charger, the main structure will use the boost power factor corrector to cascade the full-bridge power converter. The former mainly offers the power input of the high power factor, and the latter adopts the soft-switching and synchronous rectification technology in order to improve its electric conversion efficiency. For the construction of two-stage controllers, the single digital microchip controller is adopted to avoid the interference of high frequency switching of traditional power structure which uses two microchips. The experiment results have demonstrated: 1. the battery-set system can provide over two hour operation of the servant robot continuously. 2. the power management system can accurately estimates the residual capacity of battery set system. 3. the lithium-ion battery protection circuits can cut off the power apply and give the warnings when the battery set is operating under abnormal status. The battery charger allows 110V or 220V input voltage and provides the biggest 20A current to charge battery set. The power battery set can charge up to 90% capacity within one hour.

Delay-dependent fuzzy controller design for stabilization of nonlinear systems with multiple non-commensurate time delays

In this paper, the T-S fuzzy model approach is extended to design fuzzy controllers for the stabilizability of nonlinear systems with multiple non-commensurate time delays. Three types of fuzzy controllers are state feedback, observer-based state feedback, and output feedback fuzzy controller with time delay information. Based on the concept of the parallel distributed compensation (PDC) and the delay-dependent Lyapunov functional approach, some control design methods are proposed to stabilize the whole fuzzy time-delay system asymptotically. These design approaches are all dependent on time delays. By Schur complement, these sufficient conditions can be easily transformed into the problem of LMIs. The practical example based on the CSTR (continuous stirred tank reactor) model is given to illustrate the control designs and their effectiveness.