Cognition-Based Human Performance Analysis and Modeling for Collaborative Teleoperated Construction Equipment in Challenging Environments

Abstract

Excavation in real-life is a collaborative work involving several construction workers. However, the human factors which greatly affect behaviors in a team-based context are often overlooked in the excavation utility damage prevention in a challenging urban jobsite. On one hand, when operating an excavator, the excavator operator’s unsafe behavior directly affects safety accidents such as underground utility damage occurrence. On the other hand, the operator’s behavior is greatly affected by the surrounding environment (e.g., visual or auditory stimuli) and the communication with other coworkers (i.e., spotter).

To better train operators for preventing accidents in real jobsites, robotic simulators are commonly used for task practicing and studying human factors. The typical simulator is composed by joysticks and pedals, monitor-based display, and available for a single user to practice the basic operation. As such, most simulators provide an acceptable but less immersive simulation environment, and collaborative work scenario with multiple users involved have not been included in the simulation model design. Furthermore, operator’s cognitive behavior in a collaborative work environment is merely understood.

With this regard, this study lays out the fundamental work on understanding the human factors and human-robot interaction in a team-based context in challenging urban jobsite. First, this study proposes a multi-user hybrid-immersive operation and intuitive communication system in virtual reality, and the effectiveness of the proposed system was evaluated in a user experiment (N=57). Next, human factors such as operator’s performance and cognitive responses based on load theory[1] are investigated under challenging environmental conditions with different communication formats in a user experiment (N=55).