• Journal of the Korean Institute of Intelligent Systems
• /
• v.13 no.2
• /
• pp.186-191
• /
• 2003

Skills in tracing of the MR divide into following, approaching, avoiding and warning and so on. It is difficult to have all these skills learned as neural network. To make this up for, skills consisted of each module, and Mobile Robot was controlled by the output of module adequate for the situation. A mobile Robot was equipped multi-ultrasonic sensor and a USB Camera, which can be in place of human sense, and the measured environment information data is learned through Modular Neural Network. MNN consisted of optimal combination of activation function in the Expert Network and its structure seemed to improve learning time and errors. The Gating Network(GN) used to control output values of the MNN by switching for angle and speed of the robot. In the paper, EN of Modular Neural network was designed optimal combination. Traveling with a real MR was performed repeatedly to verity the usefulness of the MNN which was proposed in this paper. The robot was properly controlled and driven by the result value and the experimental is rewarded with good fruits.

• Journal of IKEEE
• /
• v.16 no.3
• /
• pp.235-243
• /
• 2012

In this study, we developed two legged multi-joint robot by using wireless servo motor that was applied by wireless sensor network technology, which is widely used recently, and performed an experiment of walking method of two legged multi-joint robot. We constructed the star network with servo motors which were used at each joint of two-legged robot. And we designed the robot for operation by transmission of joint control signal from main control system or by transmission of the status of each joint to the main control system, so it operates with continuously checking the status of joints at same time. We developed the humanoid robot by using wireless digital servo motor which is different from existing servo motor control system, and controlled it by transmitting the information of angles and speeds of robot joints to the motor(node) as a feedback through main control system after connecting power and setting up the IDs to each joint. We solved noisy problem generated from wire and wire length to connection point of the control device by construction of the wireless network instead of using existing control method of wiring, and also solved problem of poor real time response to gait motion by controlling the position with continuous transmission of control signals to each joint. And we found that the effective control of robot is able by performing the simulation on walking motion in advance with the developed control algorithm which was downloaded into installed memory. Also we performed the stable walking with two-legged robot by attaching pressure sensor to robot sole. And we examined the robot gait operated by application of calculated algorithm on robot movement to each joint. In this study, we studied the method of controlling robot gait motion by using wireless servo motors and measured the torque applied to each joint, and found that the developed wireless servo motor by ZigBee sensor network offers easier control of two legged robot gait and better circuit configuration of it than the existing wired control system could do.

• Journal of Institute of Control, Robotics and Systems
• /
• v.16 no.10
• /
• pp.963-968
• /
• 2010

This paper handles ZMP based control that is inspired by neural networks for humanoid robot walking on varying sloped surfaces. Humanoid robots are currently one of the most exciting research topics in the field of robotics, and maintaining stability while they are standing, walking or moving is a key concern. To ensure a steady and smooth walking gait of such robots, a feedforward type of neural network architecture, trained by the back propagation algorithm is employed. The inputs and outputs of the neural network architecture are the ZMPx and ZMPy errors of the robot, and the x, y positions of the robot, respectively. The neural network developed allows the controller to generate the desired balance of the robot positions, resulting in a steady gait for the robot as it moves around on a flat floor, and when it is descending slope. In this paper, experiments of humanoid robot walking are carried out, in which the actual position data from a prototype robot are measured in real time situations, and fed into a neural network inspired controller designed for stable bipedal walking.

• Journal of the Korean Society of Mechanical Technology
• /
• v.13 no.3
• /
• pp.7-16
• /
• 2011

This study proposes the compensation method for the mechanical deflection error of a SCARA robot. While most studies on the related subject have dealt with the development of a control algorithm for improvement of robot accuracy, this study presents the control method reflecting the mechanical deflection error which is predicted in advance. The deflection at the end of the gripper of SCARA robot is caused by the self-weights and payloads of Arm 1, Arm 2 and quill. If the deflection is constant even though robot's posture and payload vary, there may not be a big problem on robot accuracy because repetitive accuracy, that is relative accuracy, is more important than absolute accuracy in robot. The deflection in the end of the gripper varies as robot's posture and payload change. That's why the moments $M_x$, $M_y$ and $M_z$ working on every joint of a robot vary with robot's posture and payload size. This study suggests the compensation method which predicts the deflection in advance with the variations in robot's posture and payload using neural network. To do this, I chose the posture of robot and the payloads at random, found the deflections by the FEM analysis, and then on the basis of this data, made compensation possible by predicting deflections in advance successively with the variations in robot's posture and payload through neural network learning.

• Proceedings of the KIEE Conference
• /
• 2004.11c
• /
• pp.499-501
• /
• 2004

This paper deals with the implementation of a quadruped working robot using wireless sensor network with TinyOS. It is often required to install real time OS and wireless network in the mobile robot field since robots work alone without human intervention and also exchanging their information between robot systems. The suggested controller utilizes a built-in wireless network OS and makes the variance action related with human-kindly motions for a quadruped walking robot. In addition, a kinematics analysis of its structure and control architecture of robot system is suggested and verified the usefulness through the real experiment.

• Proceedings of the KIEE Conference
• /
• 2005.10b
• /
• pp.95-97
• /
• 2005

This paper deals with the implementation of walking for a humanoid robot by ZMP measurement using wireless sensor network. ZMP is measured by FSR sensors which are mounted at each corner of a sole. The wireless sensor network collects the sensor data according and exchanges robot information between host PC and a robot system. The master controller mounted on robot body receives trajectory data from the host PC via sensor network and drives the joint motor based on trajectory data. The time scheduler of the master controller controls the events at the ratio of 100ms. With this configuration, the walking of the humanoid robot KHR-1 could be realized successfully.

• Journal of Institute of Control, Robotics and Systems
• /
• v.5 no.5
• /
• pp.593-599
• /
• 1999

A mobile robot in FMS environment should be able to nevigate itself. Therefore, path planning is necessary for the mobile robot to perform its tasks without being lost. Path planning using a network model gives oprimal paths to every pair of nodes but building this model demands accurate information of environments. In this paper, a method to build a network model using sonar sensors is presented. The main idea is to build a quad tree model by using sonar sensors and convert the model to a network model for path planning. The new method has been implemented on a mobile robot. Experimental results show that the mobile robot constructs an accurate network model using inaccurate sonar data.

• Proceedings of the KIEE Conference
• /
• 2004.05a
• /
• pp.77-79
• /
• 2004

Because robot hardware architecture generally is consisted of a few sensors and motors connected to the central processing unit, this type of structure is led to time consuming and unreliable system. For analysis, one of the fundamental difficulties in real-time system is how to be bounded the time behavior of the system. When a distributed control network controls the robot, with a central computing hub that sets the goals for the robot, processes the sensor information and provides coordination targets for the joints. If the distributed system supposed to be connected to a control network, the joints have their own control processors that act in groups to maintain global stability, while also operating individually to provide local motor control. We try to analyze the architecture of network-based humanoid robot's leg part and deal with its application using the CAN(Controller Area Network) protocol.

• Journal of Institute of Control, Robotics and Systems
• /
• v.5 no.3
• /
• pp.338-343
• /
• 1999

In this paper, the fuzzy neural network for the obstacle avoidance, which consists of the straight-line navigation and the barrier elusion navigation, is proposed and examined. For the straight-line navigation, the fuzzy neural network gets two inputs, angle and distance between the line and the mobile robot, and produces one output, steering velocity of the mobile robot. For the barrier elusion navigation, four ultrasonic sensors measure the distance between the barrier and the mobile robot and provide the distance information to the network. Then the network outputs the steering velocity to navigate along the obstacle boundary. Training of the proposed fuzzy neural network is executed in a given environment in real-time. The weights adjusting uses the back-propagation of the gradient of error to be minimized. Computer simulations are carried out to examine the efficiency of the real time learning and the guiding ability of the proposed fuzzy neural network. It has been shown that the mobile robot that employs the proposed fuzzy neural network navigates more safely with and less trembling locus compared with the previous reported efforts.

• Proceedings of the KIEE Conference
• /
• 2004.05a
• /
• pp.80-82
• /
• 2004

Personal robot consist of various modules that have independent functions. Because personal robot has requirement that support various construction for user's tendency. Therefore, Middleware mechanism at support not only personal robot's main functional element but also embedded modules functional elements is required. Each module have various heterogeneous network interfaces and variable services and variables. Therefore, Middleware must support these various network interfaces. This paper, pointed in Controller Area Network(CAN) inreface that usually used in embedded system for control. For connect various heterogeneous network interfaces (Ethernet, RS232 etc..), it is necessary to modify bagic CAN frame format. And also make some kind of BUS topology for CAN network.