IMPACT OF DDA OPTIMIZATION ON MOBILE ROBOT PATH PLANNING FOR MIXED IMAGE IN IMAGE PROCESSING

: Existing problem solving in the day to environment requires computational intelligence. Path planning is one of the most important technologies in the navigation of the mobile robot, which should meet the optimization and real-time requests. The objective of the paper is to present a noble approach to find the efficient and effective path planning for mobile robot. Here first the image is located on the graph and then a quadtree is formed, according to the working space image with respect to the obstacle image. Then the NFT algorithm is used to obtain the shortest path from the start point to the goal point in the graph. Finally the DDA optimization algorithm is adopted to get the optimal path. Aiming at the shortcoming of the DDA algorithm which is easily plunging into the local minimum, DDA algorithm with NFT is put forward. The results of the simulation demonstrate the effectiveness of the proposed method, which can meet the real-time requests of the mobile robot's navigation. Here we have taken two different types of image, one square shape and other is mixed image of different shapes like triangle and circle. The working space is tested and result is verified using NFT Algorithm with DDA optimization.


I. INTRODUCTION
The area home land robotics has assumed a greater importance in the present age and robots are now used extensively to rescue survivors from dangerous environments when dealing with hazardous substances. Here the substance is taken as image and the image is presented as obstacle [2]. The goal of the path planning method is to determine a sequence of configurations for the robot to move around obstacles and avoid collisions while reaching a desired goal [11]. The Digital Differential Analyzer (DDA) method is widely used for planning the path of mobile robot. In a mobile robot path planning researchers used many algorithms for optimization and DDA is one of them and is considered to be subjected to other methods. This has two sections: first one is storing the location points in a vector array and second one is resolving the array step wise. In the DDA field method, we can imagine that all obstacles are represented by an image and we applied quadtree method on the image to make a tree [1]. once the tree is formed we applied the algorithm .The distance from the robot to obstacles will be judge on the basis of the tree structure. While the destination has followed with NFT (Neighbor finding technique) . Then we apply DDA on the path which is stored in the data base it has specific function and is finally the line of path is drawn. The function slopes down towards the target point, so that the robot can reach the target by following the path. We have applied this DDA algorithm on different shape or mixed image obstacles as shown in fig 9 and fig 10.we have applied DDA to 50 different locations and tested in c++ , here we have shown 10 different locations and as in the table III, and plotted the graph as shown in fig 13 and fig 14.

II. THE APPROACH
We have implemented DDA on A*, NFT algorithm to get the optimization .It is simulated and verified the result .The detail of the result and graph is shown in figure 3.

A. A*Algorithm
A* algorithm based implementation is easier and practically faster .to reach the destination, A*algorithm creates sub optimal paths using its neighbors. In A* representation, f'(n) = g(n)+h'(n), where g(n) is the total distance from the initial position to the current position and h'(n) is the estimated distance from the current position to the goal destination/state. To create this estimation a heuristic function is used. f'(n) is the sum of g(n) and h'(n) and is stated as the current estimated shortest path.

A. Application of DDAa optimization
The DDA optimization network resolve the task to reach all the nodes on the map. The route from the starting node to the target node is resolved, but the solution is not optimal or close to the optimal different methods have been adopted to get the optimized path , DDA is one of them. The proposed DDA Algorithm for finding a closely optimal route between the starting and the ending node is presented in the following: • Elimination of the duplicate nodes. If the number of neurons is equal to the number of nodes on the map, the network does not find the solution, because a set of neurons will not be active during the network training process. • Finding the address from the resulted array vector with the starting and ending nodes. • The Array vector values which correspond to node coordinates on the map are compared with start and end node positions. • Finding the nodes for which a neighborhood node with a lower cost to the target node exists. • Calculating the cost from the start to the end node and from the neighborhood nodes to the end node for each node found in the previous step and storing the results in a table with the following fields: current node index, neighborhood node index, and the calculated cost. • Ordering the table in ascending order according to the cost column. • Deleting the higher cost overlaps the section with a Lower cost. • Finally extracting the closely optimal path from the start to the end point.

IV. SIMULATED RESULT AND ANALYSIS
We have simulated A*, NFT algorithm with system configuration Intel® core(TM) i3-3220 GHz 3.30 and Ram-2GB (1.88 GB Usable).system type-32 -Bit Operating system. We applied DDA optimization on the oath finding and tested around 100 different locations i.e different start and goal points we fund that that the time taken and distance covered is very less ,as shown in fig 3,

IV. CONCLUSION
In this paper we have implemented DDA optimization technique on A*, NFT algorithm and tested the result taking different start and goal points. We found that the time and distance is reduced around 60 %. The amount of the existing works for each approach has been identified and classified and tested with C++ language. This paper divides the motion planning algorithms into two major groups, namely, the Conventional Approaches and Heuristic Approaches. The conventional approaches are Roadmap, grid search or Quadtree approach, here we tested in heuristic method also and the result as shown. A complete discussion of the portion of each approach in the field of robot motion planning is also presented, including different comparative figures and charts.