Comparison of Multiple Reinforcement Learning and Deep Reinforcement Learning Methods for the Task Aimed at Achieving the Goal

Abstract

Reinforcement Learning (RL) and Deep Reinforcement Learning (DRL) methods are a promising approach to solving complex tasks in the real world with physical robots. In this paper, we compare several reinforcement learning (Q-Learning, SARSA) and deep reinforcement learning (Deep Q-Network, Deep Sarsa) methods for a task aimed at achieving a specific goal using robotics arm UR3. The main optimization problem of this experiment is to find the best solution for each RL/DRL scenario and minimize the Euclidean distance accuracy error and smooth the resulting path by the Bézier spline method. The simulation and real word applications are controlled by the Robot Operating System (ROS). The learning environment is implemented using the OpenAI Gym library which uses the RVIZ simulation tool and the Gazebo 3D modeling tool for dynamics and kinematics.

Evolutionary Computation Techniques for Path Planning Problems in Industrial Robotics: A State-of-the-Art Review

“…The primary motivation of this paper lies in investigating recent advances and applications in the field of evolutionary algorithms focused on motion planning in industrial robotics, with the goal of expanding the research focus of the robotics lab Industry 4.0 Cell [29,30]. Although the utilization of evolutionary algorithms in the field of industrial robotics is growing in popularity, a text summarizing the state-of-the-art and recent developments was missing.…”

Deep Q-Learning in Robotics: Improvement of Accuracy and Repeatability

“…One of the widely used ML algorithms to control physical equipment is the deep q-learning algorithm, which belongs to the reinforcement learning ML type [ 23 ]. It allows a robot to find the best positioning accuracy through trial-and-error interactions with the environment rather than requiring positive or negative labels [ 24 , 25 , 26 , 27 , 28 , 29 ]. As a part of the general robotic operation optimization technique, it gradually finds the best positioning method and seeks to discover the most considerable cumulative reward value in each iteration There are two main advantages of using the deep q-learning algorithm in an industrial robot case: the possibility to introduce gathered live video data into it [ 30 , 31 ] and the possibility to avoid the commonly known ML problems such as overfitting.…”

Intelligent Sampling of Anterior Human Nasal Swabs using a Collaborative Robotic Arm

“…In this struggle, robotics plays the role of an assistant to medical personnel. For this reason, the Robo Medicinae I project was created to build on existing research on test samples analysis [17] at the Institute of Automation and Computer Science, Faculty of Mechanical Engineering, Brno University of Technology as part of the Industry 4.0 Cell (I4C) robotics laboratory [14,13].…”