Effect of a Training Curriculum Using Ergometer Guided by (Gemini AI) Outputs on Selected Kinematic Variables and 500m Kayak Achievement for Juniors

Abstract

The aim of this research is to solve a critical problem in kayak training: the problem of diagnosing and correcting small technical errors in a kinematic open water environment, which is not possible with naked eye observation for young paddlers to capture the high-speed kinematic details. The researcher wanted to design a “hybrid” training program for kayak students that integrates the stability of the Kayak Ergometer with the real-time analysis and diagnosis of the Gemini AI model. The researcher employed a matched groups approach (control and experimental). The sample of students were (10) young kayakers from the training center of Iraqi Canoe Federation. The experimental group took the proposed curriculum for 8 weeks and used Gemini AI to analyze the video footage and provide immediate visual feedback to correct motor paths on the ergometer to the correct movement patterns while the control group continued with the traditional training. We see that the experimental group outperformed the control group in the post-tests with the 500m achievement time dramatically (7.40 seconds faster than the control group). Also we found significantly improving kinematic variables (Distance Per Stroke - DPS, Stroke Rate, and Stroke Index). The researcher points out that AI and land-based training together will help to 'automate' correct movement patterns and reduce the 'cost' of effort. The study suggests implementing smart analysis tools as a fundamental part of junior training programs.