Pedestrian is the main Vulnerable Road User (VRU) that needs further support when designing enhanced Intelligent Transportation Systems (ITSs). The pedestrian’s perception and motion identification in real time and in different situations are still challenging tasks for the autonomous vehicle designer. Serious efforts have been invested in this field using recent Artificial Intelligence (AI) techniques such as deep and reinforcement learning. However, pedestrian refuge island perception and decision-making have not been investigated so far. In this paper, we propose a model to detect pedestrians and count their numbers in refuge islands using the Pedestrian Location Perception Network (P-LPN) technique. Also, our model uses the Long Short-Term Memory (LSTM) for updating the number of pedestrians over time and predict the desired decision.
Deep Learning-Based Modeling of Pedestrian Perception and Decision-Making in Refuge Island for Autonomous Driving
EAI/Springer Innovations in Communication and Computing
2022-02-21
12 pages
Article/Chapter (Book)
Electronic Resource
English
Parameterized Decision-making with Multi-modal Perception for Autonomous Driving
ArXiv | 2023
|Deep Reinforcement Learning in Decision-Making of Autonomous Driving: A Survey
BASE | 2023
|