LARIAT: Predictive Haptic Feedback to Improve Semi-Autonomous UGV Safety in a Case Study

Area of Science:

• Robotics
• Human-Computer Interaction
• Control Systems

Background:

• Advancing autonomous capabilities in unmanned ground vehicles (UGVs) is ongoing, but some applications still require human oversight.
• Limitations exist in the effectiveness of human operators for teleoperation, necessitating improved human-machine interfaces.
• Haptic feedback is a promising method to enhance operator awareness and control in semi-autonomous systems.

Purpose of the Study:

• To present the experimental validation of LARIAT (Lowering Attention Requirements in semi-Autonomous Teleoperation), a portable haptic device.
• To demonstrate LARIAT's capability to improve teleoperation performance and safety in semi-autonomous UGVs.
• To assess the impact of haptic feedback on operator attention and UGV stability.

Main Methods:

• Design and implementation of the LARIAT portable haptic device.
• Utilization of an adapted predictive Zero-Moment Point (ZMP) rollover index for haptic cue generation.
• Experimental characterization of the device's just noticeable difference and a case study simulating UGV operation.

Main Results:

• LARIAT effectively provides haptic squeeze cues to the operator based on rollover risk.
• Experimental validation confirmed the device's functionality and operator feedback thresholds.
• A case study showed LARIAT reduced UGV rollovers by up to 50% in a simulated teleoperation task.

Conclusions:

• LARIAT significantly enhances the safety and performance of semi-autonomous UGV teleoperation.
• Haptic feedback, as implemented by LARIAT, can effectively reduce critical failures like rollovers.
• The system demonstrates a viable approach for maintaining human-on-the-loop control in complex robotic applications.