Social Zone as a Barrier Function for Socially-Compliant Robot Navigation

TL;DR

This study addresses the challenge of integrating social norms into robot navigation, which is essential for ensuring that robots operate safely and efficiently in human-centric environments by utilizing the comprehensive ATC dataset to identify the minimum social zones humans and robots must respect.

Abstract

This study addresses the challenge of integrating social norms into robot navigation, which is essential for ensuring that robots operate safely and efficiently in human-centric environments. Social norms, often unspoken and implicitly understood among people, are difficult to explicitly define and implement in robotic systems. To overcome this, we derive these norms from real human trajectory data, utilizing the comprehensive ATC dataset to identify the minimum social zones humans and robots must respect. These zones are integrated into the robot's navigation system by applying barrier functions, ensuring the robot consistently remains within the designated safety set. Simulation results demonstrate that our system effectively mimics human-like navigation strategies, such as passing on the right side and adjusting speed or pausing in constrained spaces. The proposed framework is versatile, easily comprehensible, and tunable, demonstrating the potential to advance the development of robots designed to navigate effectively in human-centric environments.

Figures (6)

• Figure 1: Pedestrian trajectories in the ATC shopping mall. The central square has a large open space and exhibits low pedestrian density, which is appropriate for investigating human interactions.
• Figure 2: Examples of processed trajectories of two-person interactions from the open space. Since the data has been collected over an extended period, we can obtain trajectory scenarios of two individuals encountering each other from various speeds and directions. Although this shows 200 example trajectories, we have gathered a total of 16,181 trajectories.
• Figure 3: All distances based on the LOS angle derived from the discrete trajectories of two individuals. This roughly indicates the minimum social distance required for each angle of encounter. Given that this data comes from real-world observations, it may contain noise and outliers. Our goal is to establish the minimum boundary for the majority of the data.
• Figure 4: The distance data according to the reference pedestrian's speed, along with the dataset's outliers and enclosing convex hull. The distance dataset is represented through the complementary distance $r'$ to eliminate outliers effectively.
• Figure 5: Estimated minimum social zones according to pedestrian speed from the ATC dataset.