This work presents an ACC-like longitudinal controller for an autonomous electric vehicle, named Ego-Vehicle, based on a Deep Deterministic Reinforcement Learning algorithm. More specifically, the designed algorithm exploits the use of the Deep Deterministic Policy Gradient (DDPG) agent and the reward function explicitly takes into account both the speed and position error of the Ego-Vehicle w.r.t. the preceding one. After properly training the DDPG agent, the control ACC-like strategy is validated considering a realistic driving cycle for the preceding vehicle. Numerical results confirm the effectiveness of the designed strategy.

Adaptive Cruise Control for Autonomous Electric Vehicles based on Q-learning algorithm

Coppola A.;Petrillo A.;Rizzo R.;Santini S.
2021

Abstract

This work presents an ACC-like longitudinal controller for an autonomous electric vehicle, named Ego-Vehicle, based on a Deep Deterministic Reinforcement Learning algorithm. More specifically, the designed algorithm exploits the use of the Deep Deterministic Policy Gradient (DDPG) agent and the reward function explicitly takes into account both the speed and position error of the Ego-Vehicle w.r.t. the preceding one. After properly training the DDPG agent, the control ACC-like strategy is validated considering a realistic driving cycle for the preceding vehicle. Numerical results confirm the effectiveness of the designed strategy.
978-88-87237-50-4
File in questo prodotto:
Non ci sono file associati a questo prodotto.

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11588/880113
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 0
  • ???jsp.display-item.citation.isi??? ND
social impact