Abstract—This paper deals with the design and the implementation of an automatic task planner for a robot, irrespective of whether it is a stationary robot or a mobile robot. The aim of the task planner nothing but, they are planning systems which are used to plan a particular task and do the robotic manipulation. This planning system is embedded into the system software in the computer, which is interfaced to the computer. When the instructions are given using the computer, this is transformed into real time application using the robot. All the AI based algorithms are written and saved in the control software, which acts as the intelligent task planning system. Keywords—AI, Robot, Task Planner, RT, Algorithm, Specs, Controller. I. INTRODUCTION NE of the most important problems in robotics is the task- planning problem. A task is a job or an application or an operation that has to be done by the robot, whether it is a stationary robot or a mobile robot. The word planning means deciding on a course of action before acting. Before a robot does a particular task, how the task has to be done or performed in its workspace has to be planned. This is what is called as Robot Task Planning [2]. The tasks will be usually specified by the user and is therefore considered as a problem that has to be given to the task planner [1]. A plan is a representation of a course of action for achieving the goal. How the problem has to be solved has to be planned properly. Robot task planning is also called as problem solving techniques and is one of the important topics of Artificial Intelligence. For example, when a problem is given to a human being to be solved; first, he or she thinks about how to solve the problem, and then devises a strategy or a plan how to tackle the problem. Then only he or she starts solving the problem. Hence, robot task planning is also called as robot problem solving techniques. Many of the sub-topics in this chapter are currently under active research in the fields of Artificial Intelligence, Image Processing and Robotics. Dr. T.C. Manjunath has a Ph.D. from the renowned Systems and Control Engineering Department of the Indian Institute of Technology Bombay, Post-graduate from LD College, Gujarat Univeristy & Graduate from RVCE, Bangalore University. E-mail: tcmanjunath@rediffmail.com, tcmanjunath@gmail.com. Cemal Ardil is with the National Academy of Aviation, AZ 1056 Baku, Azerbaijan. Lot of research is going on in the robot problem solving techniques. A typical robot problem solving consists of doing a household work; say, opening a door and passing through various doors to a room to get a object. Here, it should take into consideration, the various types of obstacles that come in its way; also the front image of the scene has to be considered the most. Hence, robot vision plays a important role. In a typical formulation of a robot problem, we have a robot that is equipped with an array of various types of sensors and a set of primitive actions that it can perform in some easy way to understand the world. Robot actions change one state or configuration of one world into another. For example, there are several labeled blocks lying on a table and are scattered. A robot arm along with a camera system is also there. The task is to pick up these blocks and place them in order [4]. In a majority of the other problems, a mobile robot with a vision system can be used to perform various tasks in a robot environment containing other objects such as to move objects from one place to another ; i.e., doing assembly operations avoiding all the collisions with the obstacles [3]. The paper is organized in the following sequence. A brief introduction about the task planner and the task planning system was given in the previous paragraphs. The section gives information about the task planner design. Section 3 gives the information about the inputs-outputs of the task planner. Section 4 gives information about the brief explanation of the design of the task planner. This is followed by the conclusions in section 5 and the references. II. PHYSICAL SYSTEM DESIGN & A BRIEF REPORT The mechanical design of the developed system is divided into 3 parts, viz., the base assembly, the arm assembly and the gripper assembly. The designed robot has R-R-P (Rotary- Rotary-Prismatic) type of axes [2]. A four axis / four DOF designed SCARA robot arm as shown in Fig. 1. A SCARA robot is a 4 DOF stationary robot arm having base, elbow, vertical extension and tool roll and consisting of both rotary and prismatic joints. There is no yaw and pitch, only roll. There are 4 joints, 4 axis (three major axes - base, elbow, vertical extension and one minor axis - tool roll) [4]. The 4 DOF’s are given by Base, Elbow, Vertical Extension and Roll, i.e., there are three rotary joints and one prismatic joint. Since n = 4; 16 Design of an Artificial Intelligence Based Automatic Task Planner or a Robotic System T. C. Manjunath, C. Ardil O World Academy of Science, Engineering and Technology International Journal of Computer and Systems Engineering Vol:1, No:6, 2007 1846 International Scholarly and Scientific Research & Innovation 1(6) 2007 ISNI:0000000091950263 Open Science Index, Computer and Systems Engineering Vol:1, No:6, 2007 publications.waset.org/10008322/pdf