Suppressing Control Oscillations in Reinforcement Learning via External Action Filtering for Autonomous Driving

Osman Ozturan M. Y., Dirik A. E.

2025 16th International Conference on Electrical and Electronics Engineering, ELECO 2025, İstanbul, Turkey, 27 - 29 November 2025, (Full Text)

  • Publication Type: Conference Paper / Full Text
  • Doi Number: 10.1109/eleco69582.2025.11329257
  • City: İstanbul
  • Country: Turkey
  • Bursa Uludag University Affiliated: Yes

Abstract

Deep Reinforcement Learning (DRL) agents operating in continuous action spaces for autonomous driving often exhibit high-frequency, non-human-like control oscillations, which compromise passenger comfort, predictability and social acceptance. This paper addresses this challenge by proposing a modular DRL architecture that decouples the primary control policy from the secondary objective of action smoothing. The framework features a Soft Actor-Critic (SAC) agent trained for a combined lane-keeping and static obstacle-avoidance task. The agent's raw outputs are processed by a dedicated, external filtering module before actuation. A systematic evaluation demonstrates that a Weighted Moving Average (WMA) filter reduces action volatility by over 92% compared to an unfiltered baseline, achieving exceptionally smooth, human-like control with only a minimal compromise in task accuracy. This result validates the decoupled approach as a robust and flexible paradigm for developing socially acceptable and comfortable autonomous driving systems.