U-NEXT: A User-Centered New Prototype for Walking Excavator Simulator.

Area of Science:

• Engineering
• Human-Computer Interaction
• Virtual Reality

Background:

• Walking excavators present significant control, instruction, and safety challenges.
• Virtual reality (VR) offers a promising avenue for training complex machinery operators.
• Integrating real hardware with physics simulation enhances VR training realism.

Purpose of the Study:

• To develop and evaluate a VR walking excavator simulator.
• To investigate the impact of user perspectives (first-person vs. third-person) on performance and preference.
• To assess the effectiveness of video passthrough for controller representation in VR training.

Main Methods:

• Implementation of a VR walking excavator simulator with integrated hardware and physics.
• User study 1: 45 participants evaluated performance and preference across different user perspectives.
• User study 2: 12 participants compared video passthrough controllers with animated controllers.

Main Results:

• Task-dependent perspectives are optimal; third-person view excelled in complex leg movement tasks.
• Third-person view aided planning and visualization but did not reduce motion sickness or improve training preference.
• Video passthrough significantly enhanced controller learning and reduced errors compared to animated controllers.

Conclusions:

• VR simulators can effectively address walking excavator training challenges.
• Design recommendations favor task-specific perspective selection and video passthrough for controller representation.
• These findings inform the development of more effective VR-based training systems for heavy machinery.