Swing Evaluation System Using Sensor Data Classification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing swing evaluation methods for baseball and softball players fail to accurately assess swing characteristics in response to different types of pitches and conditions, limiting the ability to improve player skills and develop effective hitting tools.
Innovation Solution
A method and system that utilize sensors attached to a ball hitting tool and the player's body to calculate and classify swing information based on motion characteristics of the ball, such as speed, type of pitch, and course, allowing for the extraction of representative data to evaluate and compare swing characteristics across various conditions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If swing evaluation is performed using only general speed measurement without considering ball motion characteristics, then the evaluation system is simple, but the evaluation precision is insufficient for different pitch types
Solution Approach 1:
The patent segments the evaluation system into multiple functional modules: sensor data acquisition unit, swing information calculation unit, classification unit, extraction unit, and evaluation unit. Each module handles a specific aspect of the evaluation process, allowing the system to achieve high precision for different pitch types while maintaining manageable complexity through modular design.
Solution Approach 2:
The patent changes the evaluation parameters from simple speed measurement to comprehensive swing information including swing timing, speed, acceleration, rotational speed, radius of rotation, vertical bat angle, and swing orbit. These parameters are adjusted and classified based on ball motion characteristics such as pitch type, speed, and course, enabling precise evaluation across different pitch types.
2Adaptability or versatility
If swing information is collected without classification based on ball motion characteristics, then data collection is straightforward, but the ability to provide targeted training advice is limited
Solution Approach 1:
The patent implements dynamic classification where swing information is grouped based on real-time ball motion characteristics including pitch type, speed, and course. The classification criteria can be adjusted dynamically to match different game situations and training objectives, allowing the system to adapt to various evaluation needs without requiring complete system redesign.
Solution Approach 2:
The patent introduces an intermediary data processing layer that includes the classification unit and extraction unit. This intermediary layer processes raw sensor data and swing information before final evaluation, acting as a bridge between data collection and evaluation. It enables targeted training advice by extracting representative swing information from classified groups without requiring direct complex processing of all raw data.
3Measurement precision
If multiple sensors are attached to capture comprehensive swing information, then measurement accuracy improves, but the ease of operation and setup becomes more difficult
Solution Approach 1:
The patent employs sensors with multi-functionality that can capture multiple types of data simultaneously. The sensor data acquisition unit is designed to collect comprehensive swing information including position, speed, acceleration, and orientation using integrated sensor arrays, reducing the need for multiple specialized sensors and simplifying the attachment process.
Solution Approach 2:
The patent merges multiple sensor functions into integrated measurement systems. By combining various sensing capabilities into unified sensor units attached to the ball hitting tool and subject, the system achieves comprehensive measurement accuracy while reducing the number of separate attachment points and simplifying setup procedures.
Data Source
AI summary
An evaluation method for evaluating a swing characteristic of a subject with respect to a flying and thus coming object, includes: calculating swing information corresponding to a motion characteristic of the object based on sensor data detected by at least one of a first sensor attached to a ball hitting tool or a back of a hand of the subject and a second sensor attached to a waist of the subject when the subject swings the ball hitting tool with respect to the object; classifying a plurality of pieces of such swing information into a plurality of groups based on the motion characteristic of the object; for each of the plurality of groups, extracting representative swing information in the group based on one or more pieces of swing information belonging to the group; and evaluating a swing characteristic of the subject based on the representative swing information of each group.