Reinforcement learning is a well-proven and powerful algorithm for robotic arm manipulation. There are various applications of this in healthcare, such as instrument assisted surgery and other medical interventions where surgeons cannot find the target successfully. Reinforcement learning is an area of machine learning and artificial intelligence that studies how an agent should take actions in an environment so as to maximize its total expected reward over time. It does this by trying different ways through trial-and-error, hoping to be rewarded for the results it achieves. The focus of this paper is to use a deep reinforcement learning neural network to map the raw pixels from a camera to the robot arm control commands for object manipulation.
In complex planning and control operations and tasks like manipulating objects, assisting experts in various fields, navigating outdoor environments, and exploring uncharted territory, modern robots are designed to complement or completely replace humans. Even for those skilled in robot programming, designing a control schema for such robots to carry out these tasks is typically a challenging process that necessitates starting from scratch with a new and distinct controller for each task. The designer must consider the wide range of circumstances the robot might encounter. This kind of manual programming is typically expensive and time consuming. It would be more beneficial if a robot could learn the task on its own rather than having to be preprogrammed to perform all these tasks. In this paper, a method for the path planning of a robot in a known environment is implemented using Q-Learning by finding an optimal path from a specified starting and ending point.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.