Developing and Evaluating a Mixed Sensor Smart Chair System for Real-Time Posture Classification: Combining Pressure...

Area of Science:

• Ergonomics and Human Factors
• Biomedical Engineering
• Occupational Health

Background:

• Prolonged sitting and poor posture contribute to musculoskeletal disorders in workers.
• Accurate real-time posture monitoring is crucial for ergonomic interventions.
• Existing posture monitoring systems have limitations in accuracy and cost.

Purpose of the Study:

• To develop and evaluate a novel sensor-embedded smart chair system for real-time worker posture classification.
• To assess the utility of combining pressure and distance sensors for improved posture monitoring.
• To compare the performance of the mixed sensor system against single-sensor benchmark systems.

Main Methods:

• A mixed sensor system combining six pressure sensors (seat cushion) and six infrared reflective distance sensors (seatback) was developed.
• A k-Nearest Neighbor algorithm was employed to classify instantaneous sitting postures.
• Posture classification performance was evaluated against two benchmark systems using single sensor types.

Main Results:

• The mixed sensor smart chair system demonstrated significantly superior posture classification performance compared to benchmark systems.
• The system accurately classifies postures relevant to ergonomic analyses of seated work.
• The system is low-cost and utilizes a small number of sensors.

Conclusions:

• The novel mixed sensor smart chair system offers an accurate and cost-effective solution for real-time posture monitoring.
• This technology has potential applications in developing real-time posture feedback systems.
• The system can aid in the prevention of sitting-related musculoskeletal disorders.