步態為人類步行的行為特徵，泛指人在行走過程中肢體協調運作的表現。人類行走的步態是透過大腦控制經由各個神經元傳遞至骨骼肌上的神經元，產生骨骼肌的收縮達到對應關節的運動。足底地面反作用力（Ground Reaction Force, GRF）和壓力中心軌跡（Center of Pressure, COP）是評估步態表現的重要特徵，其中COP為行走過程中腳部與地面接觸於特定時間內所產生壓力的合力位置，可提供人體步態平衡、下肢肌肉神經控制能力以及反應步態功能性的優點。過去對於步態意圖的預測或分類皆為使用腦電波（Electroencephalogram, EEG）來進行，不過EEG收集數據不易且訊號複雜，而步態地面反作用力及足底壓力中心數據可連續收集且較易分析，因此本研究欲透過步態之GRF和COP資訊分類步態起始意圖。
本研究使用測力板、腦波儀以及表面肌肉電位圖（Electromyography, EMG）用以收集地面反作用力、壓力中心、腦電波與小腿脛前肌肉活動電位於步態起始之相關數據。研究對象為10名健康之成年男性，進行三種狀態，包含靜止、直走和右轉起始步態資訊收集。探討步態起始意圖各部位起始時間點的關聯性，與利用GRF與COP進行LSTM模型分類步態起始意圖的三種狀態包含靜止、直行以及右轉向。
研究結果顯示直行與右轉向之腦電波於步態起始意圖產生時間點具有統計顯著差異；COP發生位移變化時間點也具有統計顯著差異。於分類步態起始意圖之結果顯示使用GRF與COP作為特徵值，依據擺動腳離地程度進行LSTM模型分類，其平均準確度最高可達94.79%。由結果可知可以透過步態起始之GRF與COP資訊，可分類步態起始意圖。未來研究可針對不同的族群並增加樣本多樣性，提供更多步態起始意圖特徵作為訓練集以及使用不同的分類、辨識模型與不同的特徵類別進行各個模型表現績效的比較。在實際應用方面，可以透過將可量測步態參數之感測元件內嵌於行走表面，以擷取步態起始意圖之足底壓力中心與地面反作用力特徵，來達到步態意圖辨識效果，此應用對於環境的偵測與系統監控安全能更便捷的使用。

Gait is the behavioral characteristic of human walking, which generally refers to the performance of human limbs during walking. The brain controls human locomotion through the neurons transmitted to the bones by each neuron, which produces the contraction of the skeletal muscles to achieve the movement of the corresponding joints. During the gait initiation, the brain transmits the information, and then the muscles of the legs receive and complete the movement. Ground reaction force (GRF) and center of pressure (Center of Pressure, COP) are the characteristics of evaluating gait performance, where COP is the pressure produced by the contact between the foot and the ground during a specific period of time during walking. The combined force position can provide the body's gait balance, lower limb muscle and nerve control ability, and the advantages of gait function. In the past study, the prediction or classification of gait intentions used brain waves (Electroencephalogram, EEG) to predict and classify gait intentions. However, it is not easy to collect data from EEG and the signals are complex. In contrast, ground reaction force and center of pressure are easy to collect and analyze. Hence, the present study aims to classify the initial intention of gait through the GRF and COP information.
In the experiment, the force plate was used to measure ground reaction force and calculate the center of pressure. While electroencephalogram (EEG) was to measure brain waves and electromyography (EMG) was to detect the anterior tibial muscle activity of the calf and to determine the timing of gait initiation. Ten healthy young male adults were conducted in three conditions, including standstill, walking straight, and turning right for the experiment of gait intention. The timing of gait intention would be evaluated by EEG and the initiation of COP. Furthermore, GRF and COP would be treated as features to classify the gait intention about standstill, walking straight, and turning right in the LSTM model.
The results revealed that the timings of brain waves and the COP displacement initiation for gait intention were statistically significant differences between the conditions of walking straight and turning right. For the classification, the average accuracy of the LSTM model with GRF and COP as feature reached the highest one, 94.79%, depended on the heel- or toe-off of the swing leg. The results indicated that gait intentions could be classified based on the GRF and COP in gait. Future research can target different ethnic groups, increase sample diversity, provide more gait intention features as training sets, and use different classifications, identification models, and different feature categories to compare the performance of each model. In practical applications, sensors can be embedded on the walking surface to measure the center of pressure and ground reaction force for features of gait intention recognition. This application is more convenient to use for monitoring systems and environment security.

摘要 I
Abstract III
目錄 V
表目錄 VII
圖目錄 VIII
第一章 緒論 1
1.1. 研究背景與動機 1
1.2. 研究目的與範圍 4
1.3. 研究架構與流程 4
第二章 文獻回顧 6
2.1. 步態起始特性 6
2.1.1. 步態起始動作之相關研究 6
2.1.2. 步態起始意圖特徵 9
2.1.3. 步態時間性參數切割方式 10
2.2. 大腦與步態起始之關聯性 16
2.2.1. 影響步態起始之相關因素 16
2.2.2. 大腦影響步態之研究 17
2.3. 步態起始意圖與腦波之關聯性 18
2.4. 基於步態之起始意圖辨識 21
2.4.1. 步態起始意圖預測 22
2.5. 小結 24
第三章 研究方法 26
3.1. 問題定義與描述 26
3.2. 實驗設置 26
3.2.1. 實驗流程 26
3.2.2. 研究對象與實驗設備 26
3.2.3. 實驗方法與步驟 29
3.3. 數據分析 31
3.3.1. 數據同步處理方式 31
3.3.2. 壓力中心 32
3.3.3. 步態切割 33
3.3.4. 事件相關去同步（ Event Related Desynchronization, ERD） 36
3.3.5. 數值正規化 38
3.3.6. 統計分析 38
3.4. 機器學習 38
3.4.1. 分類策略 39
3.5. 前測數據 41
第四章 實驗結果 43
4.1. 受測者資訊 43
4.2. 步態起始時間點關聯性 43
4.3. 步態起始意圖分類之結果 48
第五章 討論 54
5.1. 步態起始意圖時間點之表現 54
5.2. 步態起始意圖分類結果 56
5.3. 研究限制 60
第六章 結論與未來方向 62
參考文獻 64
附錄 68

Alam, M. N., Garg, A., Munia, T. T. K., Fazel-Rezai, R., & Tavakolian, K. (2017). Vertical ground reaction force marker for Parkinson’s disease. PLoS One, 12(5), e0175951.
2. Alharthi, A. S., & Ozanyan, K. B. (2019). Deep learning for ground reaction force data analysis: Application to wide-area floor sensing. 2019 IEEE 28th International Symposium on Industrial Electronics (ISIE),
3. Annweiler, C., Beauchet, O., Bartha, R., Wells, J. L., Borrie, M. J., Hachinski, V., & Montero-Odasso, M. (2013). Motor cortex and gait in mild cognitive impairment: a magnetic resonance spectroscopy and volumetric imaging study. Brain, 136(3), 859-871.
4. Assaiante, C., Woollacott, M., & Amblard, B. (2000). Development of postural adjustment during gait initiation: kinematic and EMG analysis. Journal of motor behavior, 32（3), 211-226.
5. Baars, B. J., & Gage, N. M. （2010). Cognition, brain, and consciousness: Introduction to cognitive neuroscience. Academic Press.
6. Bai, O., Rathi, V., Lin, P., Huang, D., Battapady, H., Fei, D.-Y., . . . Hallett, M. (2011). Prediction of human voluntary movement before it occurs. Clinical neurophysiology, 122(2), 364-372.
7. Boiten, F., Sergeant, J., & Geuze, R. (1992). Event-related desynchronization: the effects of energetic and computational demands. Electroencephalography and clinical neurophysiology, 82(4), 302-309.
8. Caderby, T., Yiou, E., Peyrot, N., Begon, M., & Dalleau, G. (2014). Influence of gait speed on the control of mediolateral dynamic stability during gait initiation. Journal of biomechanics, 47(2), 417-423.
9. Chen, L.-X. (2020). Gender and age recognition via center of pressure in gait.
10. Chiu, M.-C., Wu, H.-C., & Chang, L.-Y. (2013). Gait speed and gender effects on center of pressure progression during normal walking. Gait & posture, 37(1), 43-48.
11. Delorme, A., & Makeig, S. (2004). EEGLAB: an open source toolbox for analysis of single-trial EEG dynamics including independent component analysis. Journal of neuroscience methods, 134(1), 9-21.
12. Derlatka, M., & Bogdan, M. (2015). Ensemble kNN classifiers for human gait recognition based on ground reaction forces. 2015 8th International Conference on Human System Interaction (HSI),
13. Dipaola, M., Pavan, E. E., Cavallari, P., & Frigo, C. A. (2017). Anticipatory postural adjustments of gait initiation in healthy subjects: The effect of interfoot distance. 2017 E-Health and Bioengineering Conference (EHB),
14. Elble, R. J., Moody, C., Leffler, K., & Sinha, R. (1994). The initiation of normal walking. Movement disorders, 9(2), 139-146.
15. Graves, A. (2008). Supervised Sequence Labelling with Recurrent Neural Networks [Ph. D. dissertation]. Technical University of Munich, Germany.
16. Hasan, S. S., Siddiquee, M. R., & Bai, O. (2020). Asynchronous Prediction of Human Gait Intention in a Pseudo Online Paradigm Using Wavelet Transform. IEEE Transactions on Neural Systems and Rehabilitation Engineering, 28(7), 1623-1635.
17. Hochreiter, S., & Schmidhuber, J. (1997). Long short-term memory. Neural computation, 9(8), 1735-1780.
18. Khanmohammadi, R., Talebian, S., Hadian, M. R., Olyaei, G., & Bagheri, H. (2016). Characteristic muscle activity patterns during gait initiation in the healthy younger and older adults. Gait & posture, 43, 148-153.
19. Klimesch, W., Doppelmayr, M., Pachinger, T., & Russegger, H. (1997). Event-related desynchronization in the alpha band and the processing of semantic information. Cognitive Brain Research, 6(2), 83-94.
20. Lee, J. M., & Kim, C. H. (2019). Detection of Gait Intention with an Insole Device. 2019 IEEE Biomedical Circuits and Systems Conference (BioCAS),
21. Lelard, T., Doutrellot, P.-L., Temfemo, A., & Ahmaidi, S. (2017). Electromyographic pattern during gait initiation differentiates yoga practitioners among physically active older subjects. Frontiers in human neuroscience, 11, 300.
22. Leocani, L., Toro, C., Manganotti, P., Zhuang, P., & Hallett, M. (1997). Event-related coherence and event-related desynchronization/synchronization in the 10 Hz and 20 Hz EEG during self-paced movements. Electroencephalography and Clinical Neurophysiology/Evoked Potentials Section, 104(3), 199-206.
23. Mansouri, N., Issa, M. A., & Jemaa, Y. B. (2017). Gait features fusion for efficient automatic age classification. IET Computer Vision, 12(1), 69-75.
24. Mason, J. E., Traoré, I., & Woungang, I. (2016). Machine Learning Techniques for Gait Biometric Recognition. Springer.
25. Millett, D. (2001). Hans Berger: From psychic energy to the EEG. Perspectives in biology and medicine, 44(4), 522-542.
26. Moustakidis, S. P., Theocharis, J. B., & Giakas, G. (2008). Subject recognition based on ground reaction force measurements of gait signals. IEEE Transactions on Systems, Man, and Cybernetics, Part B (Cybernetics), 38(6), 1476-1485.
27. Novak, D., Reberšek, P., Beravs, T., Podobnik, J., Munih, M., De Rossi, S. M. M., . . . Carrozza, M. C. (2012). Early recognition of gait initiation and termination using wearable sensors. 2012 4th IEEE RAS & EMBS International Conference on Biomedical Robotics and Biomechatronics (BioRob),
28. Ortiz, M., Rodríguez-Ugarte, M., Iáñez, E., & Azorín, J. M. (2017). Application of the Stockwell transform to electroencephalographic signal analysis during gait cycle. Frontiers in neuroscience, 11, 660.
29. Paranjape, R., Mahovsky, J., Benedicenti, L., & Koles, Z. (2001). The electroencephalogram as a biometric. Canadian Conference on Electrical and Computer Engineering 2001. Conference Proceedings (Cat. No. 01TH8555),
30. Park, S., Choi, H., Ryu, K., Kim, S., & Kim, Y. (2009). Kinematics, kinetics and muscle activities of the lower extremity during the first four steps from gait initiation to the steady-state walking. Journal of Mechanical Science and Technology, 23(1), 204-211.
31. Perry, J., & Davids, J. R. (1992). Gait analysis: normal and pathological function. Journal of Pediatric Orthopaedics, 12(6), 815.
32. Pfurtscheller, G., & Da Silva, F. L. (1999). Event-related EEG/MEG synchronization and desynchronization: basic principles. Clinical neurophysiology, 110(11), 1842-1857.
33. Planelles, D., Hortal, E., Costa, A., Iáñez, E., & Azorín, J. M. (2014). First steps in the development of an EEG-based system to detect intention of gait initiation. 2014 IEEE International Systems Conference Proceedings,
34. Qian, G., Zhang, J., & Kidané, A. (2008). People identification using gait via floor pressure sensing and analysis. European Conference on Smart Sensing and Context,
35. Reberšek, P., Novak, D., Podobnik, J., & Munih, M. (2011). Intention detection during gait initiation using supervised learning. 2011 11th IEEE-RAS International Conference on Humanoid Robots,
36. Sahyoun, C., Floyer-Lea, A., Johansen-Berg, H., & Matthews, P. (2004). Towards an understanding of gait control: brain activation during the anticipation, preparation and execution of foot movements. Neuroimage, 21(2), 568-575.
37. Samson, M. M., Crowe, A., De Vreede, P., Dessens, J. A., Duursma, S. A., & Verhaar, H. J. (2001). Differences in gait parameters at a preferred walking speed in healthy subjects due to age, height and body weight. Aging clinical and experimental research, 13(1), 16-21.
38. Saripalle, S. K., Paiva, G. C., Cliett III, T. C., Derakhshani, R. R., King, G. W., & Lovelace, C. T. (2014). Classification of body movements based on posturographic data. Human movement science, 33, 238-250.
39. Sburlea, A. I., Montesano, L., de la Cuerda, R. C., Diego, I. M. A., Miangolarra-Page, J. C., & Minguez, J. (2015). Detecting intention to walk in stroke patients from pre-movement EEG correlates. Journal of NeuroEngineering and Rehabilitation, 12(1), 1-12.
40. Schneiders, A. G., Sullivan, S. J., O'Malley, K. J., Clarke, S. V., Knappstein, S. A., & Taylor, L. J. (2010). A valid and reliable clinical determination of footedness. PM&R, 2(9), 835-841.
41. Segal, A. D., Orendurff, M. S., Czerniecki, J. M., Shofer, J. B., & Klute, G. K. (2008). Local dynamic stability in turning and straight-line gait. Journal of biomechanics, 41(7), 1486-1493.
42. Segel, J. D., & Crawford, S. (2014). Anatomy of the COP gait line and computer-aided gait analysis. Pm’s Tech Forum/Orthotics & Biomechanics,
43. Serrao, M., Ranavolo, A., Andersen, O. K., Conte, C., Don, R., Cortese, F., . . . Sandrini, G. (2012). Adaptive behaviour of the spinal cord in the transition from quiet stance to walking. BMC neuroscience, 13(1), 1-15.
44. Shafiul Hasan, S., Siddiquee, M. R., Atri, R., Ramon, R., Marquez, J. S., & Bai, O. (2020). Prediction of gait intention from pre-movement EEG signals: a feasibility study. Journal of NeuroEngineering and Rehabilitation, 17, 1-16.
45. Soriano-Segura, P., Iáñez, E., Quiles, V., Ferrero, L., Ortiz, M., & Azorín, J. M. (2020). Selection of Spatial, Temporal and Frequency Features to Detect Direction Changes During Gait. 2020 42nd Annual International Conference of the IEEE Engineering in Medicine & Biology Society (EMBC),
46. Strike, S. C., & Taylor, M. J. (2009). The temporal–spatial and ground reaction impulses of turning gait: is turning symmetrical? Gait & posture, 29(4), 597-602.
47. Suzuki, M., Miyai, I., Ono, T., & Kubota, K. (2008). Activities in the frontal cortex and gait performance are modulated by preparation. An fNIRS study. Neuroimage, 39(2), 600-607.
48. Vrieling, A. H., Van Keeken, H., Schoppen, T., Otten, E., Halbertsma, J., Hof, A., & Postema, K. (2008). Gait initiation in lower limb amputees. Gait & posture, 27(3), 423-430.
49. Wong, L., Hunt, A., Burns, J., & Crosbie, J. (2008). Effect of foot morphology on center-of-pressure excursion during barefoot walking. Journal of the American Podiatric Medical Association, 98(2), 112-117.
50. Woo, J., Ho, S., Lau, J., Chan, S., & Yuen, Y. (1995). Age-associated gait changes in the elderly: pathological or physiological? Neuroepidemiology, 14（2), 65-71.
51. Yiou, E., Artico, R., Teyssedre, C. A., Labaune, O., & Fourcade, P. (2016). Anticipatory postural control of stability during gait initiation over obstacles of different height and distance made under reaction-time and self-initiated instructions. Frontiers in human neuroscience, 10, 449.
52. Yiou, E., Caderby, T., Delafontaine, A., Fourcade, P., & Honeine, J.-L. (2017). Balance control during gait initiation: State-of-the-art and research perspectives. World journal of orthopedics, 8(11), 815.