Occupancy Reward-Driven Exploration with Deep Reinforcement Learning for Mobile Robot System

Abstract

This paper investigates the solution to a mobile-robot exploration problem following autonomous driving principles. The exploration task is formulated in this study as a process of building a map while a robot moves in an indoor environment beginning from full uncertainties. The sequence of robot decisions of how to move defines the strategy of the exploration that this paper aims to investigate, applying one of the Deep Reinforcement Learning methods, known as the Deep Deterministic Policy Gradient (DDPG) algorithm. A custom environment is created representing the mapping process with a map visualization, a robot model, and a reward function. The actor-critic network receives and sends input and output data, respectively, to the custom environment. The input is the data from the laser sensor, which is equipped on the robot. The output is the continuous actions of the robot in terms of linear and angular velocities. The training results of this study show the strengths and weaknesses of the DDPG algorithm for the robotic mapping problem. The implementation was developed in MATLAB platform using its corresponding toolboxes. A comparison with another exploration algorithm is also provided.

Multi-Robot Exploration of Unknown Space Using Combined Meta-Heuristic Salp Swarm Algorithm and Deterministic Coordinated Multi-Robot Exploration

“…Related studies proposed by different researchers have been based CME [ 14 ], meta-heuristics [ 37 ], and hybrid methods combining meta-heuristic with deterministic algorithms [ 38 , 39 , 40 ]. Some studies focused solely on static sensor coverage faults and robot movements in uncharted surroundings [ 41 , 42 , 43 ] as well as exploration with deep reinforcement learning for mobile robots [ 44 ]; however, they coincide in meaning but the goal of both studies is to create a finite map. The rest of the paper is organized as follows:…”

Signal Novelty Detection as an Intrinsic Reward for Robotics

“…Hyperparameter tuning was critical to the success of this method [8]. A different approach, called Occupancy-Reward-Driven Exploration [9], has been applied in robotics to explore uncharted territories within the state space. In this technique, an occupancy map is utilized to acquire information about the environment through sensors such as a laser sensor.…”

“…The robot's reward is then determined by the number of new segments discovered within the occupancy map at each time step. This approach can also improve the robot's power efficiency [9].…”