Study on Horse-Rider Interaction Based on Body Sensor Network in Competitive Equitation
Authors: Jie Li, Zhelong Wang, S. Qiu, Hongyu Zhao, Jiaxin Wang, Xin Shi, Long Liu, Ning Yang
Journal: IEEE Transactions on Affective Computing
Summary
# Editorial Summary Researchers at IEEE developed a wearable sensor system combining inertial measurement units and electroencephalography to simultaneously capture rider biomechanics and emotional state during ridden work, addressing a significant gap in objective training assessment where subjective observation has traditionally dominated. Using an extended Kalman filter algorithm to fuse mechanical and neurological data, they reconstructed rider posture across the four primary gaits and validated their motion capture accuracy against the gold-standard Vicon optical system. Key findings demonstrated that rider emotional responses varied consistently with experience level, and that kinematic patterns derived from combining postural and emotional metrics aligned well with observable riding behaviour across different competitive styles. For practitioners, this technology offers potential applications in rider development programmes by providing quantifiable biofeedback on both physical position and psychological state—enabling targeted interventions to improve seat stability, rein contact consistency and emotional regulation during training. Whilst the system currently requires bespoke IMU placement and EEG headwear, the researchers' ability to control estimation errors and cross-validate against established motion capture methods suggests this multimodal approach could eventually provide coaches, physiotherapists and instructors with objective data to complement their existing assessment tools.
Read the full abstract on the publisher's site
Practical Takeaways
- •Wearable sensor technology can objectively monitor rider position and emotional state during training, potentially identifying when fatigue or tension affects performance
- •Motion capture validation confirms that inertial sensors can reliably track postural changes across gaits without laboratory equipment, enabling field-based training analysis
- •Quantifying rider emotion alongside kinematics may help identify stress or anxiety as confounding factors in training effectiveness and horse-rider partnership quality
Key Findings
- •A body sensor network combining IMU and EEG successfully captured rider motion and emotional state during four riding styles (walk, sitting trot, rising trot, canter)
- •The exercise intensity extended Kalman filter (EID-EKF) method accurately reconstructed rider posture with errors well-controlled and validated against optical motion capture
- •Emotion changes varied between riders of different skill levels and correlated with kinematic patterns during training