The primary challenge in robotic navigation lies in enabling robots to adapt effectively to new, unseen environments. Addressing this gap, this paper enhances the Twin Delayed Deep Deterministic Policy Gradient (TD3) model's adaptability by introducing randomized start and goal points. This approach aims to overcome the limitations of fixed goal points used in prior research, allowing the robot to navigate more effectively through unpredictable scenarios. This proposed extension was evaluated in unseen environments to validate the enhanced adaptability and performance of the TD3 model. The experimental results highlight improved flexibility and robustness in the robot's navigation capabilities, demonstrating the ability of the model to generalize effectively to unseen environments. Additionally, this paper provides a concise overview of TD3, focusing on its core mechanisms and key components to clarify its implementation.
Adaptive Robot Navigation Using Randomized Goal Selection with Twin Delayed Deep Deterministic Policy Gradient / Ali, Romisaa; Dogru, Sedat; Marques, Lino; Chiaberge, Marcello. - In: ROBOTICS. - ISSN 2218-6581. - ELETTRONICO. - 14:4(2025). [10.3390/robotics14040043]
Adaptive Robot Navigation Using Randomized Goal Selection with Twin Delayed Deep Deterministic Policy Gradient
Ali, Romisaa;Chiaberge, Marcello
2025
Abstract
The primary challenge in robotic navigation lies in enabling robots to adapt effectively to new, unseen environments. Addressing this gap, this paper enhances the Twin Delayed Deep Deterministic Policy Gradient (TD3) model's adaptability by introducing randomized start and goal points. This approach aims to overcome the limitations of fixed goal points used in prior research, allowing the robot to navigate more effectively through unpredictable scenarios. This proposed extension was evaluated in unseen environments to validate the enhanced adaptability and performance of the TD3 model. The experimental results highlight improved flexibility and robustness in the robot's navigation capabilities, demonstrating the ability of the model to generalize effectively to unseen environments. Additionally, this paper provides a concise overview of TD3, focusing on its core mechanisms and key components to clarify its implementation.File | Dimensione | Formato | |
---|---|---|---|
Acceptance-Certificate-robotics-3473682.pdf
accesso riservato
Descrizione: Acceptance certificate for the article "Adaptive Robot Navigation Using Randomized Goal Selection with TD3", accepted in the MDPI Robotics journal on 25 March 2025.
Tipologia:
Altro materiale allegato
Licenza:
Non Pubblico - Accesso privato/ristretto
Dimensione
219.01 kB
Formato
Adobe PDF
|
219.01 kB | Adobe PDF | Visualizza/Apri Richiedi una copia |
Invoice_MDPI_robotics-3473682_1880.72EUR.pdf
accesso riservato
Descrizione: Invoice for article processing charges (APC) – Manuscript ID: robotics-3473682
Tipologia:
Altro materiale allegato
Licenza:
Non Pubblico - Accesso privato/ristretto
Dimensione
26.11 kB
Formato
Adobe PDF
|
26.11 kB | Adobe PDF | Visualizza/Apri Richiedi una copia |
robotics-14-00043.pdf
accesso aperto
Tipologia:
2a Post-print versione editoriale / Version of Record
Licenza:
Creative commons
Dimensione
1.94 MB
Formato
Adobe PDF
|
1.94 MB | Adobe PDF | Visualizza/Apri |
Pubblicazioni consigliate
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
https://hdl.handle.net/11583/2998601