Autonomous Vehicles that Interact with Pedestrians: A Survey of Theory and Practice (1805.11773v1)

Abstract: One of the major challenges that autonomous cars are facing today is driving in urban environments. To make it a reality, autonomous vehicles require the ability to communicate with other road users and understand their intentions. Such interactions are essential between the vehicles and pedestrians as the most vulnerable road users. Understanding pedestrian behavior, however, is not intuitive and depends on various factors such as demographics of the pedestrians, traffic dynamics, environmental conditions, etc. In this paper, we identify these factors by surveying pedestrian behavior studies, both the classical works on pedestrian-driver interaction and the modern ones that involve autonomous vehicles. To this end, we will discuss various methods of studying pedestrian behavior, and analyze how the factors identified in the literature are interrelated. We will also review the practical applications aimed at solving the interaction problem including design approaches for autonomous vehicles that communicate with pedestrians and visual perception and reasoning algorithms tailored to understanding pedestrian intention. Based on our findings, we will discuss the open problems and propose future research directions.

• The paper presents a comprehensive survey integrating pedestrian behavioral factors with AV communication methods to improve urban mobility safety.
• It analyzes social, demographic, and environmental influences on pedestrian decision-making and examines practical V2P systems.
• The study highlights the need for robust intent prediction and cross-modal communication, guiding future research in autonomous transportation.

Autonomous Vehicles that Interact with Pedestrians: A Survey of Theory and Practice

The paper "Autonomous Vehicles that Interact with Pedestrians: A Survey of Theory and Practice" by Amir Rasouli and John K. Tsotsos provides a comprehensive examination of the interaction between autonomous vehicles (AVs) and pedestrians, a critical area of research relevant to urban environments. The paper offers an in-depth analysis of pedestrian behavior, covering both historical studies involving human drivers and more recent investigations related to AV technology.

Overview and Key Findings

The research explores multiple dimensions affecting pedestrian behavior, including demographic, social, and environmental factors. It stresses that successful AV operation in urban settings necessitates sophisticated models to predict pedestrian intentions, rooted in a rich understanding of these behaviors.

Pedestrian Factors:

• Social Influences: Group size greatly impacts pedestrian decision-making, potentially manipulating AV yield tendencies. Social norms and imitation play significant roles but are influenced by cultural context.
• Demographics: Gender and age are pivotal; women and elderly individuals tend to exhibit more caution compared to men and younger pedestrians.
• Attention and Trajectory: Eye contact and walking direction are significant in assessing intentions, affecting how AVs should interpret pedestrian actions.

Environmental Factors:

• Traffic Dynamics: Gap acceptance is a critical determinant in pedestrian crossing behavior, influenced by vehicle speed, distance, and traffic flow.
• Communication Needs: The paper highlights the importance of non-verbal communication, yet it remains a polarizing area within the AV field.

Studies in Autonomous Contexts

Unlike classical studies with human drivers, pedestrian behavior around AVs is less thoroughly researched. The paper suggests that AVs may need to adopt communication modalities to bridge the interaction gap typical in human driving, such as visual or audio signals.

Contemporary Challenges:

• Technological Integration: Developing intent prediction algorithms overcome challenges found in human-AV interactions. While some studies state communication may be unnecessary, the exploration of modalities like visual displays is gaining traction.
• Data Limitations: Current datasets often lack comprehensive context, which impacts the effectiveness of machine learning models in predicting pedestrian behavior.

Practical Systems and Communication

The paper analyzes practical implementations, such as Vehicle-to-Pedestrian (V2P) communication technologies and LED/light-based signaling systems, for effective interaction with pedestrians. However, these raise privacy and social responsibility issues, necessitating further empirical evaluation.

Implications and Future Directions

The implications of this research are substantive, contributing to both practical solutions and theoretical advancements in intelligent transportation systems. The integration of pedestrian behavior models with AV systems demands rigorous testing across diverse environments to tackle current contradictions and assumptions.

Future Research Directions:

• Expanded empirical studies with larger, demographically diverse populations to validate and refine pedestrian behavior models.
• Development of unified frameworks integrating environmental context with pedestrian dynamics in AV perception systems.
• Exploration of cross-modality communication techniques to enhance AV-pedestrian interactions.

In summary, this paper serves as a keystone for understanding the complex dynamics of pedestrian interactions with AVs, leveraging insights from behavioral psychology and technological innovations. It sets a solid foundation for ongoing research crucial to enhancing the safety and efficacy of autonomous urban mobility solutions.