• 제어로봇시스템학회:학술대회논문집
• /
• 1995.10a
• /
• pp.65-68
• /
• 1995

This paper describes trajectory generation of a riobot arm by self-organizing neural networks. These neural networks are based on competitive learning without a teacher and this algorithm which is suitable for problems in which solutions as teaching signal cannot be defined-e.g. inverse dynamics analysis-is adopted to the trajectory generation problem of a robot arm. Utility of unsupervised learning algorithm is confirmed by applying the approximated solution of each joint calculated through learning to an actual robot arm in giving the experiment of tracking for reference trajectory.

• 제어로봇시스템학회:학술대회논문집
• /
• 1995.10a
• /
• pp.530-533
• /
• 1995

This paper deals with the problem of motion planning for a unicycle-like robot. We present a simple local planner for unicycle model, based on an approximation of the desired configuration generated by local holonomic planner that ignores motion constraints. To guarantee a collision avoidance, we propose an inequality constraint, based on the motion analysis with the constant control input and time interval. Consequently, we formulate our problem as the constrained optimization problem and a feedback scheme based on local sensor information is established by simply solving this problem. Through simulations, we confirm the validity and effectiveness of our algorithm.

• 제어로봇시스템학회:학술대회논문집
• /
• 1999.10a
• /
• pp.173-176
• /
• 1999

This paper shows a self tuning fuzzy inference method by the genetic algorithm in the fuzzy-sliding mode control for a Polishing robot. Using this method, the number of inference rules and the shape of membership functions are determined by the genetic algorithm. The fuzzy outputs of the consequent part are derived by the gradient descent method. Also, it is guaranteed that .the selected solution become the global optimal solution by optimizing the Akaike's information criterion expressing the quality of the inference rules. It is shown by simulations that the method of fuzzy inference by the genetic algorithm provides better learning capability than the trial and error method.

• 제어로봇시스템학회:학술대회논문집
• /
• 1999.10a
• /
• pp.292-297
• /
• 1999

Evolving in artificial agent is an extremely difficult problem, but on the other hand, a challenging task. At present the studies mainly centered on single agent learning problem. In our case, we use simulated soccer to investigate multi-agent cooperative learning. Consider the fundamental differences in learning mechanism, existing reinforcement learning algorithms can be roughly classified into two types-that based on evaluation functions and that of searching policy space directly. Genetic Programming developed from Genetic Algorithms is one of the most well known approaches belonging to the latter. In this paper, we give detailed algorithm description as well as data construction that are necessary for learning single agent strategies at first. In following step moreover, we will extend developed methods into multiple robot domains. game. We investigate and contrast two different methods-simple team learning and sub-group loaming and conclude the paper with some experimental results.

• 제어로봇시스템학회:학술대회논문집
• /
• 1990.10b
• /
• pp.1093-1097
• /
• 1990

A d.c. servomotor with pulse encoder is used to improve the movement of a hospital mobile robot along the desired line. We can achieve an improved movement of the robot by applying a PLL control. It is then shown that we can also reduce 42% of the power dissipation by the use of a PWM control. Furthermore, some simulation studies are presented to illustrate the design of PI control and optimal regulator for the control of the d.c. servomotor.

• 제어로봇시스템학회:학술대회논문집
• /
• 1986.10a
• /
• pp.49-54
• /
• 1986

In this paper we deal with a method of controlling PUMA 560 robot manipulator using a newly developed optical proximity sensor and the PDP-11/44 computer. The sensor output is different somewhat depending on an color of the object. The range of sensing vary from 38.4mm to 109.5mm for a real object, 39mm to 111.65mm for yellow, and 40.55mm to 107.25mm for blue. When an obstacle is encountered on the path of end-effector the system acknowledger immediatly the existence of the obstacle : and holds the motion of arm at a given distance. And also the system is capable of making the end effector avoid the obstacle automatically and keep on its motion.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.14 no.3
• /
• pp.1188-1203
• /
• 2020

The human expression contains a lot of information that can be used to detect complex conditions such as pain and fatigue. After deep learning became the mainstream method, the traditional feature extraction method no longer has advantages. However, in order to achieve higher accuracy, researchers continue to stack the number of layers of the neural network, which makes the real-time performance of the model weak. Therefore, this paper proposed an expression recognition framework based on densely concatenated convolutional neural networks to balance accuracy and latency and apply it to humanoid robots. The techniques of feature reuse and parameter compression in the framework improved the learning ability of the model and greatly reduced the parameters. Experiments showed that the proposed model can reduce tens of times the parameters at the expense of little accuracy.

• Journal of the Korean Society for Precision Engineering
• /
• v.3 no.1
• /
• pp.77-92
• /
• 1986

Wire harness assembly is one of the most labor intensive processes in manufa- tuing, but the process has not been fully automated yet. In this paper a variety of concepts for flexible automated assembly system are suggested to design a robot- based wire harness assembly cell. All concepts are featuring an automatic, fedkng of wires, terminating routing, inserting and bundling. Based upon possible assembly methods and procedures, six alternative systems are proposed and, finally, evaluated10 from the view point of system performance such as flexibility, reliability assembly time and equipment cost.

• 제어로봇시스템학회:학술대회논문집
• /
• 1990.10a
• /
• pp.417-422
• /
• 1990

For the robot manipulator in performing precision task, it is indispensable that the robot utilize the various sensors for intelligence. This paper presents the development and implementation of an integrated control system for the control of robotic manipulator, a feeder, a conveyor belt system, force/torque sensor system, and a photo sensor system. Micro controller board was constructed for hierarchical control of the system. To set up the program interactively, a user can make use of the software which includes the full-down menu and a dialog box. The user can make progress the program quickly and easily by a mouse. The related software was written in C and assembly languages.

• 제어로봇시스템학회:학술대회논문집
• /
• 1990.10a
• /
• pp.1-6
• /
• 1990

Neural network is a computational model of ft biological nervous system developed ID exploit its intelligence and parallelism. Applying neural networks so robots creates many advantages over conventional control methods such as learning, real-time control, and continuous performance improvement through training and adaptation. In this paper, dynamic control of a six-link robot will be presented using neural networks. The neural network model used in this paper is the backpropagation network. Simulated control of the PUMA 560 am shows that it can move a high speed as well as adapt to unforseen load changes and sensor noise. The results are compared with the conventional PD control scheme.