• Home
  • Search
  • Deep reinforcement learning multi-sensor based navigation and control
  • Cite Icon1
  • https://doi.org/10.11606/t.18.2024.tde-04062025-095959Copy DOI Icon

Deep reinforcement learning multi-sensor based navigation and control

  • Feb 23, 2024
  • Luiz Afonso Batalha Marão
Show More
  • Abstract
  • Literature Map
  • Citations
  • Similar Papers
Abstract

This thesis aims to explore a novel approach in the realm of autonomous mobile robot control. It endeavors to replace the classic mobile control structure, encompassing the control system, obstacle recognition, and avoidance, and trajectory planning, with a more advanced AI-based solution using Deep Reinforcement Learning (DRL). By embracing DRL, this research seeks to uncover innovative possibilities and evaluate its effectiveness across different difficulty levels and strategies for navigating the robots. One of the key contributions of this proposed method is its ability to perform learned data-fusion from multiple sensors, including visual inputs. By assimilating information from various sources, the AI-driven system gains a more comprehensive understanding of the environment, leading to more informed decision-making and improved navigation capabilities. To achieve this objective, a series of experiments were conducted in specially designed simulation environments, created specifically for this thesis. These carefully crafted environments, based on the Robocores Robot Trekking challenge, held in 2018 (ROBOCORE, 2018), allowed for the evaluation of different levels of complexity and required skills for successful navigation and control. The results from these experiments shed light on the trade-offs between computational resources and robots performances. As the difficulty level increases, the demand for more complex and computationally demanding resources becomes apparent to sustain a high level of performance. Additionally, the suitability of the proposed method covers diverse combinations of task complexity, severity of failure consequences, and hardware availability. The implementation of this AI-driven system using DRL showcases promising prospects for achieving more efficient and robust control of autonomous mobile robots. The findings presented in this thesis offer valuable insights for the advancement and deployment of autonomous DRL-based robotic systems in real-world scenarios. The ability to perform learned data-fusion from multi sensors (IMU, GPS and encoders) including visual inputs, consists in a significant step towards enhancing the perception and decision-making capabilities of autonomous robots in complex environments.

Similar Papers
  • Conference Article
  • Citations3

<title>CLIPS implementation of a knowledge-based distributed control of an autonomous mobile robot</title>

  • Mar 01, 1991
  • Proceedings of SPIE, the International Society for Optical Engineering/Proceedings of SPIE
  • Akram A Bou-Ghannam +1
  • Research Article
  • Citations65

On the role of stored internal state in the control of autonomous mobile robots

  • Mar 15, 1993
  • AI Magazine
  • Erann Gat
  • PDF
  • Research Article
  • Citations40

Hybrid Kalman Filter/Fuzzy Logic based Position Control of Autonomous Mobile Robot

  • Sep 01, 2005
  • International Journal of Advanced Robotic Systems
  • Rerngwut Choomuang +1
  • PDF
  • Research Article
  • Citations59

An Experimental Framework for 5G Wireless System Integration into Industry 4.0 Applications

  • Jul 23, 2021
  • Energies
  • Ignacio Rodriguez +9
  • Conference Article
  • Citations19

Autonomous Mobile Robot Control Using Fuzzy Logic and Genetic Algorithm

  • Sep 01, 2007
  • Gintautas Narvydas +2
  • Research Article

Design, Construction and Control of an Autonomous Mobile Rescue Robot with Visual Feedback

  • Jun 30, 2022
  • European Journal of Science and Technology
  • Ibrahim Hasan +1
  • Research Article

Enhancing 2D Path Search and PID Control for Autonomous Mobile Robots: A* Algorithm Optimization and Hardware

  • Feb 21, 2024
  • Advances in Engineering Technology Research
  • Chengzhi Zhou
  • Book Chapter

Research on Control System of Upper Limb Rehabilitation Training Based on Reinforcement Learning

  • Dec 06, 2022
  • Liangda Wu +4
  • Research Article

Deep Transfer Reinforcement Learning-Based Exploration Enhanced Multi-UAV Trajectory Planning

  • Nov 21, 2025
  • IEEE Transactions on Communications
  • Wenke Yuan +6
  • Book Chapter
  • Citations23

Conceptual Bases of Robot Navigation Modeling, Control and Applications

  • Jun 15, 2011
  • Silas F. R. Alves +4
  • Research Article

Research on trajectory planning based on deep reinforcement learning

  • Nov 01, 2024
  • Journal of Physics: Conference Series
  • Zan Zhou +3
  • Research Article
  • Citations15

Control of autonomous mobile robots with automated planning

  • Jan 01, 2011
  • Journal of Physical Agents (JoPha)
  • Ezequiel Quintero Barrios +3
  • Research Article
  • Citations65

Intelligent Land-Vehicle Model Transfer Trajectory Planning Method Based on Deep Reinforcement Learning.

  • Sep 01, 2018
  • Sensors
  • Lingli Yu +3
  • Research Article
  • Citations22

Visual control of autonomous mobile robot based on self‐organizing model for pattern learning

  • Oct 01, 1988
  • Journal of Robotic Systems
  • Hisashi Suzuki +1
  • Research Article
  • Citations24

A software tool for modeling and simulation of numerical P systems

  • Dec 10, 2010
  • Biosystems
  • Catalin Buiu +3
Cactus Communications logo

Copyright 2026 Cactus Communications. All rights reserved.