ABSTRACT
With the development of science and technology, robotics research has gained immense interest among various scholars. In the current state of technology use, robotics has spread to various industries such as automobiles, production, surgery, industrial automation, and so forth. In the robotics world, robots are used in various shapes and sizes; however, the humanoid form of robot is considered as the most successful form as it is capable of easing human effort in several ways and also mimicking the human behavior. In the study of humanoid robots, path planning and navigation problem is considered as one of the major aspects of design. During path planning of a humanoid robot to reach a certain goal position, problems like obstacle avoidance, goal-following behavior are encountered. To reduce the path length of travel and time taken to reach the goal position, some optimization techniques can be used. Those techniques can be based on both the classical approach and computational intelligence. In the 56current investigation, focus has been given to the use of classical approach for the navigation problem. Different classical approaches may include regression analysis, mathematical techniques, statistical approaches, and so forth. A humanoid NAO has been used for the current analysis. A regression controller has been designed for the NAO robot considering the challenges involved in the navigation problem. The regression controller is designed considering the principles of regression technique. By implementing the regression controller in the robot, simulation of several environments has been carried out. A real experimental setup has also been designed with the similar positioning of obstacles as was in the case of simulation. Parameters like path length and time taken are noted down, and finally, a comparison has been done between the simulated and experimental results. This work can also be extended toward the use of modern techniques for the same navigation problem. Dynamic environment problems such as multiple NAO navigation can also be achieved by this approach.
