Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Equimetrics -- Applying HAR principles to equestrian activities

Jonas Pöhler, Kristof Van Laerhoven

TL;DR

The paper addresses the need for objective, data-driven analysis of rider–horse performance in equestrian activities by applying human activity recognition (HAR) to a multimodal sensor setup. It introduces Equimetrics, a cost-effective open-source system using a network of wearable IMUs on both rider and horse, synchronized video, and Transformer-based HAR to classify gaits and dressage movements, as well as to extract rider-specific motion. Notable contributions include precise hoof-contact event detection with ~8.98 ms timing accuracy, high gait recognition (F1 ≈ 0.9324) and dressage task classification (F1 ≈ 0.7601), and a multimodal dataset enabling analysis of rider–horse interactions. The work demonstrates the potential for actionable, data-driven feedback to optimize training, performance, and safety, while acknowledging the need for broader validation across more horse–rider pairs and longitudinal studies.

Abstract

This paper presents the Equimetrics data capture system. The primary objective is to apply HAR principles to enhance the understanding and optimization of equestrian performance. By integrating data from strategically placed sensors on the rider's body and the horse's limbs, the system provides a comprehensive view of their interactions. Preliminary data collection has demonstrated the system's ability to accurately classify various equestrian activities, such as walking, trotting, cantering, and jumping, while also detecting subtle changes in rider posture and horse movement. The system leverages open-source hardware and software to offer a cost-effective alternative to traditional motion capture technologies, making it accessible for researchers and trainers. The Equimetrics system represents a significant advancement in equestrian performance analysis, providing objective, data-driven insights that can be used to enhance training and competition outcomes.

Equimetrics -- Applying HAR principles to equestrian activities

TL;DR

The paper addresses the need for objective, data-driven analysis of rider–horse performance in equestrian activities by applying human activity recognition (HAR) to a multimodal sensor setup. It introduces Equimetrics, a cost-effective open-source system using a network of wearable IMUs on both rider and horse, synchronized video, and Transformer-based HAR to classify gaits and dressage movements, as well as to extract rider-specific motion. Notable contributions include precise hoof-contact event detection with ~8.98 ms timing accuracy, high gait recognition (F1 ≈ 0.9324) and dressage task classification (F1 ≈ 0.7601), and a multimodal dataset enabling analysis of rider–horse interactions. The work demonstrates the potential for actionable, data-driven feedback to optimize training, performance, and safety, while acknowledging the need for broader validation across more horse–rider pairs and longitudinal studies.

Abstract

This paper presents the Equimetrics data capture system. The primary objective is to apply HAR principles to enhance the understanding and optimization of equestrian performance. By integrating data from strategically placed sensors on the rider's body and the horse's limbs, the system provides a comprehensive view of their interactions. Preliminary data collection has demonstrated the system's ability to accurately classify various equestrian activities, such as walking, trotting, cantering, and jumping, while also detecting subtle changes in rider posture and horse movement. The system leverages open-source hardware and software to offer a cost-effective alternative to traditional motion capture technologies, making it accessible for researchers and trainers. The Equimetrics system represents a significant advancement in equestrian performance analysis, providing objective, data-driven insights that can be used to enhance training and competition outcomes.
Paper Structure (7 sections, 5 figures)

This paper contains 7 sections, 5 figures.

Table of Contents

  1. Introduction
  2. Related Work
  3. Equimetrics Sensor System
  4. Preliminary Data Capture Results
  5. Automatic Analysis
  6. Activity Recognition and Classification
  7. Discussion

Figures (5)

  • Figure 1: Positions of the IMU sensors.
  • Figure 2: Highlighting the hoof-on and hoof-off events
  • Figure 3: Extraction of the principal movement of the riders leg by substraction of horse movement
  • Figure 4: The extracted movements enables the computation of activity maps for the rider
  • Figure 5: Confusion Matrix for the simple as well as for the complex classifier