[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-121589-en":3,"doc-seo-121589-105":30,"detail-sidebar-cat-0-en-105":91},{"code":4,"msg":5,"data":6},0,"success",{"doc_id":7,"user_id":8,"nickname":9,"user_avatar":10,"doc_module":4,"category_id":11,"category_name":12,"doc_title":13,"doc_description":14,"doc_content":15,"file_id":16,"file_url":17,"file_type":18,"file_size":19,"view_count":4,"is_deleted":4,"is_public":20,"is_downloadable":20,"audit_status":20,"page_count":21,"language":22,"language_code":23,"site_id":24,"html_lang":23,"table_of_contents":25,"faqs":26,"seo_title":27,"seo_description":14,"update_tm":28,"read_time":29},121589,7971461741311,"Ophelia","https://ap-avatar.wpscdn.com/avatar/74000253aff267980c6?x-image-process=image/resize,m_fixed,w_180,h_180&k=1779345379180704826",8,"Research & Report","The Role of Machine Learning in the Simulation of Body Dynamics - Bachelorarbeit","The simulation of physical body dynamics underpins robotics, engineering, and interactive systems, enabling step-by-step prediction of motion from modeled mass, inertia, and applied forces. Classical physics-based pipelines are established, yet they can face limitations in generalisation and high computational cost. The thesis surveys learning-integrated simulation methods, emphasizing forward body dynamics and comparing physics-informed learning, hybrid physics-learning models, and specialised neural architectures through accuracy, generalisation, scalability, and computational efficiency.","The Role of Machine Learning in the Simulation of Body Dynamics  \nHenry Ruß  \nMatr. Nr.: 43388  \nMedieninformatik  \nHochschule der Medien  \nSupervisor  \nProf. Dr. Roland Schmitz, Stephan Soller  \nIn partial fulfillment of the requirements for the degree of  \nBachelor of Science  \nJune 30, 2025  \nEhrenw¨ortliche Erkl¨arungen  \nHiermit versichere ich, Henry Ruß, ehrenw¨ortlich, dass ich die vorliegende Bachelorarbeit mit dem Titel: ”The role of machine learningin the simulation of body dynamics“ selbstst¨andig und ohne fremde Hilfe verfasst und keine anderen als die angegebenen Hilfsmittel benutzt habe. Die Stellen der Arbeit, die dem Wortlaut oder dem Sinn nach anderen Werken entnommen wurden, sind in jedem Fall unter Angabe der Quelle kenntlich gemacht. Ebenso sind alle Stellen, die mit Hilfe eines KI-basierten Schreibwerkzeugs erstellt oder ¨uberarbeitet wurden, kenntlich gemacht. Die Arbeit ist noch nicht ver¨offentlicht oder in anderer Form als Pr¨ufungsleistung vorgelegt worden.  \nIch habe die Bedeutung der ehrenw¨ortlichen Versicherung und die pr¨ufungsrechtlichen Folgen (§ 24 Abs. 2 Bachelor-SPO, § 23 Abs. 2 Master-SPO (Vollzeit)) einer unrichtigen oder unvollst¨andigen ehrenw¨ortlichen Versicherung zur Kenntnisgenommen.  \nAbstract  \nThe simulation of physical body dynamics plays a central role in robotics, engineering, and interactive systems. Classical physics-based methods are well established, but they often struggle with limited generalisation or high computational cost. In response, recent research has explored how machine learning techniques can enhance or replace traditional simulation pipelines. This thesis presents a structured overview of current methods that integrate learning into physical simulation, focusing specifically on forward simulation of body dynamics. Three methodological paradigms are examined: (i) physics-informed machine learning, which embed physical principles directly into the learning process; (ii) hybrid models, which combine classical physics engines with learned components; (iii) and specialised neural architectures, including graph-based and recurrent models, tailored to capture object dynamics. For each category, representative approaches from the literature are analysed in terms of accuracy, generalisation, scalability, and computational efficiency. The findings highlight the potential of machine learning to improve simulation quality, accelerate computation, and broaden applicability. However, they also reveal persistent challenges such as stability in long rollouts, the need for high-quality data, and limited benchmarking standards across domains.  \nContents  \nList of Acronyms v  \n1 Introduction 1  \n2 Motivation 4  \n2.1 Relevance of Physical Simulations in Modern Applications ..... 4  \n2.2 Challenges of Classic Simulation Methods .............. 5  \n2.3 Potential of Machine Learning in Physics Simulations ........ 6  \n3 Methods 9  \n3.1 Scope and Rationale .......................... 9  \n3.2 Source Selection and Literature Identification ............ 10  \n3.3 Data Organisation and Categorisation ................ 11  \n3.4 Evaluation and Differentiation Criteria ................ 11  \n4 Fundamentals 13  \n4.1 Machine learning ............................ 13  \n4.1.1 Neural networks ......................... 14  \n4.1.2 Optimisation algorithms .................... 20  \n4.1.3 Learning Strategies ....................... 20  \n4.1.4 Important Terms ........................ 21  \n4.2 Body Dynamics ............................. 22  \n4.2.1 Body Types ........................... 22  \n4.2.2 Application Domains ...................... 24  \n4.2.3 Forward vs. Inverse Dynamics ................. 26  \n4.3 Physic Simulation for Body Dynamics ................. 27  \n4.3.1 Simulators ............................ 28  \n4.3.2 Differentiable simulators .................... 30  \n5 Approaches 32  \n5.1 Physics Informed Machine Learning .................. 32  \n5.1.1 Neural Ordinary Differential Equations ............ 34 ","cbCaif03Y9DPh7D0","https://ap.wps.com/l/cbCaif03Y9DPh7D0","pdf",859255,1,74,"English","en",105,"# Introduction\n## Motivation\n## Relevance of Physical Simulations in Modern Applications\n## Challenges of Classic Simulation Methods\n## Potential of Machine Learning in Physics Simulations\n# Methods\n## Scope and Rationale\n## Source Selection and Literature Identification\n## Data Organisation and Categorisation\n## Evaluation and Differentiation Criteria\n# Fundamentals\n## Machine learning\n## Body Dynamics\n## Physic Simulation for Body Dynamics\n# Approaches\n## Physics Informed Machine Learning\n## Hybrid Models\n## Specialised Neural Networks\n# Conclusion\n# Future Work","[{\"question\":\"What problem does the thesis address in body dynamics simulation?\",\"answer\":\"It addresses how classical physics-based simulation pipelines can struggle with limited generalisation and high computational cost, motivating the integration of machine learning into physical simulation for forward dynamics.\"},{\"question\":\"Which main methodological paradigms are examined?\",\"answer\":\"The thesis examines physics-informed machine learning, hybrid models that combine physics engines with learned components, and specialised neural architectures such as graph-based and recurrent models tailored to object dynamics.\"},{\"question\":\"How are different approaches evaluated and compared?\",\"answer\":\"Approaches are analysed using criteria such as accuracy, generalisation, scalability, and computational efficiency, with representative literature methods mapped to each category.\"}]","The Role of Machine Learning in the Simulation of Body Dynamics - Bachelorarbeit | PDF",1785736378,186,{"code":4,"msg":31,"data":32},"ok",{"site_id":24,"language":23,"slug":33,"title":13,"keywords":34,"description":14,"schema_data":35,"social_meta":86,"head_meta":88,"extra_data":90,"updated_unix":28},"the-role-of-machine-learning-in-the-simulation-of-body-dynamics-bachelor-thesis","",{"@graph":36,"@context":85},[37,54,68],{"@type":38,"itemListElement":39},"BreadcrumbList",[40,44,48,51],{"item":41,"name":42,"@type":43,"position":20},"https://docshare.wps.com","Home","ListItem",{"item":45,"name":46,"@type":43,"position":47},"https://docshare.wps.com/document/","Document",2,{"item":49,"name":12,"@type":43,"position":50},"https://docshare.wps.com/document/research-report/",3,{"item":52,"name":13,"@type":43,"position":53},"https://docshare.wps.com/document/the-role-of-machine-learning-in-the-simulation-of-body-dynamics-bachelor-thesis/121589/",4,{"url":52,"name":13,"@type":55,"author":56,"headline":13,"publisher":58,"fileFormat":61,"inLanguage":23,"description":14,"dateModified":62,"datePublished":62,"encodingFormat":61,"isAccessibleForFree":63,"interactionStatistic":64},"DigitalDocument",{"name":9,"@type":57},"Person",{"url":41,"name":59,"@type":60},"DocShare","Organization","application/pdf","2026-08-03",true,{"@type":65,"interactionType":66,"userInteractionCount":4},"InteractionCounter",{"@type":67},"ViewAction",{"@type":69,"mainEntity":70},"FAQPage",[71,77,81],{"name":72,"@type":73,"acceptedAnswer":74},"What problem does the thesis address in body dynamics simulation?","Question",{"text":75,"@type":76},"It addresses how classical physics-based simulation pipelines can struggle with limited generalisation and high computational cost, motivating the integration of machine learning into physical simulation for forward dynamics.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"Which main methodological paradigms are examined?",{"text":80,"@type":76},"The thesis examines physics-informed machine learning, hybrid models that combine physics engines with learned components, and specialised neural architectures such as graph-based and recurrent models tailored to object dynamics.",{"name":82,"@type":73,"acceptedAnswer":83},"How are different approaches evaluated and compared?",{"text":84,"@type":76},"Approaches are analysed using criteria such as accuracy, generalisation, scalability, and computational efficiency, with representative literature methods mapped to each category.","https://schema.org",{"og:url":52,"og:type":87,"og:title":13,"og:site_name":59,"og:description":14},"article",{"robots":89,"canonical":52},"index,follow",{"doc_id":7,"site_id":24},{"code":4,"msg":5,"data":92},[93,97,101,105,110,115,120,123,128,131,135],{"id":20,"doc_module":4,"doc_module_name":46,"category_name":94,"show_sort_weight":95,"slug":96},"Story & Novel",90,"story-novel",{"id":47,"doc_module":4,"doc_module_name":46,"category_name":98,"show_sort_weight":99,"slug":100},"Literature",80,"literature",{"id":53,"doc_module":4,"doc_module_name":46,"category_name":102,"show_sort_weight":103,"slug":104},"Exam",70,"exam",{"id":106,"doc_module":4,"doc_module_name":46,"category_name":107,"show_sort_weight":108,"slug":109},5,"Comic",60,"comic",{"id":111,"doc_module":4,"doc_module_name":46,"category_name":112,"show_sort_weight":113,"slug":114},6,"Technology",50,"technology",{"id":116,"doc_module":4,"doc_module_name":46,"category_name":117,"show_sort_weight":118,"slug":119},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":121,"slug":122},30,"research-report",{"id":124,"doc_module":4,"doc_module_name":46,"category_name":125,"show_sort_weight":126,"slug":127},9,"Religion & Spirituality",20,"religion-spirituality",{"id":126,"doc_module":4,"doc_module_name":46,"category_name":129,"show_sort_weight":126,"slug":130},"World Cup","world-cup",{"id":132,"doc_module":4,"doc_module_name":46,"category_name":133,"show_sort_weight":132,"slug":134},10,"Lifestyle","lifestyle",{"id":136,"doc_module":4,"doc_module_name":46,"category_name":137,"show_sort_weight":106,"slug":138},19,"General","general"]