English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

This study presents a comparative analysis between single-objective andmulti-objective reinforcement learning methods for training a robot to navigateeffectively to an end goal while efficiently avoiding obstacles. Traditionalreinforcement learning techniques, namely Deep Q-Network (DQN), DeepDeterministic Policy Gradient (DDPG), and Twin Delayed DDPG (TD3), have beenevaluated using the Gazebo simulation framework in a variety of environmentswith parameters such as random goal and robot starting locations. These methodsprovide a numerical reward to the robot, offering an indication of actionquality in relation to the goal. However, their limitations become apparent incomplex settings where multiple, potentially conflicting, objectives arepresent. To address these limitations, we propose an approach employingMulti-Objective Reinforcement Learning (MORL). By modifying the reward functionto return a vector of rewards, each pertaining to a distinct objective, therobot learns a policy that effectively balances the different goals, aiming toachieve a Pareto optimal solution. This comparative study highlights thepotential for MORL in complex, dynamic robotic navigation tasks, setting thestage for future investigations into more adaptable and robust roboticbehaviors.

Enhancing Robotic Navigation: An Evaluation of Single and Multi-Objective Reinforcement Learning Strategies