Exploring AI Integration in Robotics for Adaptive Navigation Technique and Enhanced Utility.

Ruyel Rodrigues

Dr. Lili Ma and Dr. Benito Mendoza

The objective of this study was to incorporate Artificial Intelligence into the locomotive behavior of the robot and observe the impact it has on the robot’s performance. Robotic autonomy is typically constrained by static programmed behavior, whereas an AI-driven robot is better equipped for adaptive evolution in its actions. It is important for a robot that is deployed in the real world to navigate uncertain terrain because it will frequently face obstacles that are unknown. The process started by selecting an appropriate Linux operating system and the corresponding Robot Operating System (ROS) conducive to AI algorithm development. After running simulations that checked the robots environment, the robots sensor and motor functions.  The machine learning program was formed and executed to test and examine the robot’s approach to navigate to its goal. Series of trials with slightly adjusted features were performed to fine tune the algorithm. With a reliable algorithm established, the environments were updated to introduce complexity in room design and new obstacles. The findings in this research demonstrated that the robot that navigated with artificial intelligence had a significantly better chance of reaching its goal than a typical statically programmed robot. Due to the nature of robots without Ai, it is impossible to train it without encountering the same obstacles repeatedly, something that the Ai driven robot learned to adapt to with each trial. The result is a robot that appears to have some level of awareness of itself and its surroundings that is much better equipped for real world deployment. In an effort to create robots that support human lives and contribute to the betterment of our livelihood, these findings demonstrate a step towards achieving that goal using Ai in robotics.