在棒球中，記錄投球過程和投球參數需要昂貴的設備和後續處理。在本篇論文中，我們設計了一個智慧棒球，將我們設計的晶片嵌入到棒球當中。我們的開發版是以 MSP432P401R 作為主要微控制器，CC2640R2F 作為藍牙低功耗控制器，快閃記憶體作為內部儲存的裝置。在外圍傳感器部分，我們使用ICM-20649作為慣性測量單元和薄膜的力感測器A301來感測手指力。我們記錄慣性測量單元和指力數據，並通過藍牙低功耗模組將數據傳輸到電腦或是手機。之後，我們在電腦上進行信號處理，計算出球速、旋轉速度、投球軌跡和手指力。

我們提供五種不同的操作模式，包括兩顆慣性測量單元的感測模式、最高取樣率模式、即時傳輸模式、力感測模式。用戶可以選擇符合自己要求的傳感器配置。另外還新增了一個按下力感測器來啟動六軸感測器高取樣率的模式並將數據存儲在快閃記憶體中。它可以連續使用長達 24 小時。在感應範圍的部分，我們可以測量到球速為每秒120公里，旋轉速度為每分鐘2000轉，手指力為25磅。

在本論文中，我們將加速度取樣率從 1125赫茲提高到 4500赫茲。取樣率的提高可以提高球速和旋轉速率的計算精度。至此，我們完成了一次投球實驗。實驗分為兩部分。一個使用1125赫茲的取樣率和10次投擲來計算最大球速，另一部分使用4500赫茲的取樣率也進行10次投擲來計算最大球速。第一部分的均方根誤差率與測速槍相比為百分之4.876。第二部分的均方根誤差率與測速槍相比為百分之4.281。我們進行了獨立樣本T檢定來檢測兩個實驗在統計上的誤差，可以看到兩組實驗的誤差在統計上是沒有顯著差異的，可以得知在球速的計算上不需要過高的取樣率。

In baseball, recording the pitching process and pitching parameters requires expensive equipment and subsequent processing. In this thesis, we designed a smart baseball that embedded our evaluation broad inside the baseball. The evaluation broad used MSP432P401R as the main microcontroller, CC2640R2F as the Bluetooth low energy (BLE) controller, and NAND Flash as the storage component. At the parts of the peripheral sensors, we used ICM-20649 as the inertial measurement unit (IMU) and A301 to sense the finger force. We record the IMU and force data and transmit the data via the BLE module to the personal computer (PC). After that, we conduct the signal processing on a PC and calculate the ball speed, spin rate, pitching swing trajectory, and finger force.

We provide five different operation modes, including double IMU sensing mode, the highest sample rate mode, a real-time transmission mode, a force detect mode, and an output storage mode. Users can choose the sensor configuration which meets their requirements. In addition, we can directly press the force sensor to activate the high sampling rate mode and store the data in the flash memory. It can be used continuously for up to 24 hours. In the parts of the sensing range, we can measure the ball speed up to 120 kilometers per second, spin rate up to 2000 rpm, and finger force up to 25 pounds.

In this thesis, we improve the acceleration sample rate from 1.125 kHz to 4.5 kHz. The sample rate improvement can improve the ball speed and spin rate calculation accuracy. Thus, we completed a pitching experiment. The experiment was divided into two parts. One used the sampling rate of 1.125 kHz and pitched 10 times to calculate the maximum ball speed, and the other part used the sampling rate of 4.5 kHz also pitched 10 times to calculate the maximum ball speed. In the first part, the RMSE error rate was 4.876% compared with the radar gun. In the second part, the RMSE error rate was 4.281% compared with the radar gun. We completed the independent sample t-test of the error between both experiments. There is no statistically significant difference between the two experiments. A higher sampling rate is not necessary for the calculation of ball speed.

[1] M. Lapinski, E. Berkson, T. Gill, M. Reinold, and J. A. Paradiso, “A distributed wearable, wireless sensor system for evaluating professional baseball pitchers and batters,” International Symposium on Wearable Computers, pp. 131–138, 2009.
[2] H. KODA, K. SAGAWA, K. KUROSHIMA, T. TSUKAMOTO, K. URITA, and
Y. ISHIBASHI, “3d measurement of forearm and upper arm during throwing motion using body mounted sensor,” Journal of Advanced Mechanical Design, Systems, and Manufacturing, vol. 4, no. 1, pp. 167–178, 2010.
[3] Y.-W. Bai and I.-H. Hsieh, “Using a wearable device to assist the training of the throwing motion of baseball players,” IEEE Canadian Conference of Electrical and Computer Engineering (CCECE), pp. 1–6, 2019.
[4] M. Lapinski, C. Brum Medeiros, D. Moxley Scarborough, E. Berkson, T. J. Gill, T. Kepple, and J. A. Paradiso, “A wide-range, wireless wearable inertial motion sensing system for capturing fast athletic biomechanics in overhead pitching,” Sensors, vol. 19, no. 17, 2019. [Online]. Available: https://www.mdpi.com/1424-8220/19/17/3637
[5] N. Murray, G. Black, R. Whiteley, P. Gahan, M. Cole, A. Utting, and T. Gabbett, “Automatic detection of pitching and throwing events in baseball with inertial measurement sensors,” International Journal of Sports Physiology and Performance, vol. 12, pp. 1–18,09 2016.
[6] R. S. McGinnis and N. C. Perkins, “A highly miniaturized, wireless inertial measurement unit for characterizing the dynamics of pitched baseballs and softballs,” Sensors, vol. 12, no. 9, pp. 11 933–11 945, 2012. [Online]. Available:https://www.mdpi.com/1424-8220/12/9/11933
[7] H. Kinoshita, S. Obata, D. Nasu, K. Kadota, T. Matsuo, and G. S. Fleisig, “Finger forces in fastball baseball pitching,” Human movement science, vol. 54, p. 172—181, August 2017. [Online]. Available: https://doi.org/10.1016/j.humov.2017.04.007
[8] R. S. McGinnis, N. C. Perkins, and K. King, “Pitcher training aided by instrumented baseball,” Procedia Engineering, vol. 34, pp. 580–585, 2012, eNGINEERING OF SPORT CONFERENCE 2012. [Online]. Available: https: //www.sciencedirect.com/science/article/pii/S1877705812017122
[9] Jingletek, “Jingletek strike sales kit,” August 2018.
[10] ImpactTM Owner’s Manual, GARMIN, May 2018, rev. A.
[11] CASIO Digital Camera EX-F1 User’s Guide, CASIO, Jan 2008.
[12] Rapsodo, “Rapsodo pitching user manual,” April 2019. [Online]. Available: https://rapsodo.com/wp-content/uploads/2019/04/User Manual Pitching v6.pdf
[13] BALL COACHTM QUICK START GUIDE, Pocket Radar, May 2014.
[14] T. Nagami, T. Higuchi, and K. Kanosue, “How baseball spin influences the performance of a pitcher,” The Journal of Physical Fitness and Sports Medicine, vol. 2, no. 1, pp. 63–68, 2013.
[15] World’s First Wide-Range 6-Axis MEMS MotionTrackingTM Device for Sports and High Impact Applications, InvenSense, Dec. 2016, rev. 1.0.
[16] BHI160 Ultra-low power sensor hub incl. integrated IMU, Bosch Sensortec, Jul. 2016, rev. 1.0.
[17] FlexiForce®Standard Model A301, Tekscan, 2016, rev. F.
[18] MSP432P401R, MSP432P401M SimpleLink Mixed-Signal Microcontrollers, Texas Instruments,, Mar. 2015, [Revised Mar 2017].
[19] CC2640R2F SimpleLink Bluetooth low energy Wireless MCU, Texas Instruments, Dec 2016, [Revised Jan 2017].
[20] GD5F2GQ4xF DATASHEET, GigaDevice,, Mar. 2018, rev. 2.0.
[21] Orientalmotor BMU Manual, Orientalmotor, Sep 2019.
[22] T. Pfau and P. Reilly, “How low can we go? influence of sample rate on equine pelvic displacement calculated from inertial sensor data,” Equine Veterinary Journal, vol. 53, no. 5, pp. 1075–1081, 2021.
[23] T. A. Marusenkova, “An algorithm for detecting the minimum allowable accelerometer sample rate for tracing translational motion along a b´ezier curve,” , no. 1(85), p. 147–154, 2020. [Online]. Available: http://ten.ztu.edu.ua/article/view/206078
[24] Series 2200 Multichannel Programmable DC Power Supplies Declassification and Security Instructions, Keithley Instruments, Dec 2013, rev. B.
[25] Model 2100 6 1/2-Digit Resolution Digital Multimeter Calibration Manual, Keithley Instruments, Mar 2021, rev. C.
[26] C. H. Chen, “A smart sensing baseball.” Master’s thesis, National Tsing Hua University, Hsinchu, Taiwan, 2020.
[27] ADXL377: Small, Low Power, 3-Axis ±200 g Accelerometer Data Sheet, Analog devices, Apr. 2019, rev. B.
[28] ADXRS649: Fast Starting, ±20,000°/sec Vibration Rejecting Rate Gyro Data Sheet, Analog devices, Oct. 2012, rev. B.