Pub Date : 2024-12-02DOI: 10.1146/annurev-control-022723-033252
Alberta Longhini, Yufei Wang, Irene Garcia-Camacho, David Blanco–Mulero, Marco Moletta, Michael C. Welle, Guillem Alenyà, Hang Yin, Zackory Erickson, David Held, Júlia Borràs, Danica Kragic
The realm of textiles spans clothing, households, healthcare, sports, and industrial applications. The deformable nature of these objects poses unique challenges that prior work on rigid objects cannot fully address. The increasing interest within the community in textile perception and manipulation has led to new methods that aim to address challenges in modeling, perception, and control, resulting in significant progress. However, this progress is often tailored to one specific textile or a subcategory of these textiles. To understand what restricts these methods and hinders current approaches from generalizing to a broader range of real-world textiles, this review provides an overview of the field, focusing specifically on how and to what extent textile variations are addressed in modeling, perception, benchmarking, and manipulation of textiles. We conclude by identifying key open problems and outlining grand challenges that will drive future advancements in the field.
{"title":"Unfolding the Literature: A Review of Robotic Cloth Manipulation","authors":"Alberta Longhini, Yufei Wang, Irene Garcia-Camacho, David Blanco–Mulero, Marco Moletta, Michael C. Welle, Guillem Alenyà, Hang Yin, Zackory Erickson, David Held, Júlia Borràs, Danica Kragic","doi":"10.1146/annurev-control-022723-033252","DOIUrl":"https://doi.org/10.1146/annurev-control-022723-033252","url":null,"abstract":"The realm of textiles spans clothing, households, healthcare, sports, and industrial applications. The deformable nature of these objects poses unique challenges that prior work on rigid objects cannot fully address. The increasing interest within the community in textile perception and manipulation has led to new methods that aim to address challenges in modeling, perception, and control, resulting in significant progress. However, this progress is often tailored to one specific textile or a subcategory of these textiles. To understand what restricts these methods and hinders current approaches from generalizing to a broader range of real-world textiles, this review provides an overview of the field, focusing specifically on how and to what extent textile variations are addressed in modeling, perception, benchmarking, and manipulation of textiles. We conclude by identifying key open problems and outlining grand challenges that will drive future advancements in the field.","PeriodicalId":29961,"journal":{"name":"Annual Review of Control Robotics and Autonomous Systems","volume":"8 1","pages":"295-322"},"PeriodicalIF":0.0,"publicationDate":"2024-12-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147382173","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-07-01Epub Date: 2023-11-21DOI: 10.1146/annurev-control-062823-024352
Amal Nanavati, Vinitha Ranganeni, Maya Cakmak
Over 1 billion people in the world are estimated to experience significant disability. These disabilities can impact people's ability to independently conduct activities of daily living, including ambulating, feeding, dressing, taking care of personal hygiene, and more. Mobile and manipulator robots, which can move about human environments and physically interact with objects and people, have the potential to assist people with disabilities in activities of daily living. Although the vision of physically assistive robots has motivated research across sub-fields of robotics for decades, such robots have only recently become feasible in terms of capabilities, safety, and price. More and more research involves end-to-end robotic systems that interact with people with disabilities in real world settings. In this paper, we survey papers about physically assistive robots intended for people with disabilities from top conferences and journals in robotics, human-computer interactions, and accessible technology, to identify the general trends and research methodologies. We then dive into three specific research themes - interaction interfaces, levels of autonomy, and adaptation - and present frameworks for how these themes manifest across physically assistive robot research. We conclude with directions for future research.
{"title":"Physically Assistive Robots: A Systematic Review of Mobile and Manipulator Robots that Physically Assist People with Disabilities.","authors":"Amal Nanavati, Vinitha Ranganeni, Maya Cakmak","doi":"10.1146/annurev-control-062823-024352","DOIUrl":"10.1146/annurev-control-062823-024352","url":null,"abstract":"<p><p>Over 1 billion people in the world are estimated to experience significant disability. These disabilities can impact people's ability to independently conduct activities of daily living, including ambulating, feeding, dressing, taking care of personal hygiene, and more. Mobile and manipulator robots, which can move about human environments and physically interact with objects and people, have the potential to assist people with disabilities in activities of daily living. Although the vision of physically assistive robots has motivated research across sub-fields of robotics for decades, such robots have only recently become feasible in terms of capabilities, safety, and price. More and more research involves end-to-end robotic systems that interact with people with disabilities in real world settings. In this paper, we survey papers about physically assistive robots intended for people with disabilities from top conferences and journals in robotics, human-computer interactions, and accessible technology, to identify the general trends and research methodologies. We then dive into three specific research themes - interaction interfaces, levels of autonomy, and adaptation - and present frameworks for how these themes manifest across physically assistive robot research. We conclude with directions for future research.</p>","PeriodicalId":29961,"journal":{"name":"Annual Review of Control Robotics and Autonomous Systems","volume":"7 ","pages":"123-147"},"PeriodicalIF":14.0,"publicationDate":"2024-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12978752/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147445019","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-11-14DOI: 10.1146/annurev-control-060923-025701
Mrdjan Jankovic
Operating autonomous agents in unstructured space presents a difficult problem. The complexity of making decisions such as when to yield and when to go ahead increases exponentially with the number of agents. This is true for humans as well as for software that controls autonomous agents. With some practice, however, human operators are able to move efficiently in a maze of interacting agents in dense traffic. One recent result correlates the instability of equilibria in a multiagent system with an absence of gridlocks. These control barrier function–based algorithms do not include a decision-making component—the action is continuous, and negotiation happens through instability. This mechanism, referred to as instinctive negotiation, is contrasted with discontinuity-induced decisions arising from nonconvex optimization. Based on observed behavioral similarities and insights into human implicit and explicit learning, this article proposes a connection with human driving and suggests that humans may employ a mechanism similar to instinctive negotiation to navigate dense traffic. Expected final online publication date for the Annual Review of Control, Robotics, and Autonomous Systems, Volume 7 is May 2024. Please see http://www.annualreviews.org/page/journal/pubdates for revised estimates.
{"title":"Instinctive Negotiation by Autonomous Agents in Dense, Unstructured Traffic: A Controls Perspective","authors":"Mrdjan Jankovic","doi":"10.1146/annurev-control-060923-025701","DOIUrl":"https://doi.org/10.1146/annurev-control-060923-025701","url":null,"abstract":"Operating autonomous agents in unstructured space presents a difficult problem. The complexity of making decisions such as when to yield and when to go ahead increases exponentially with the number of agents. This is true for humans as well as for software that controls autonomous agents. With some practice, however, human operators are able to move efficiently in a maze of interacting agents in dense traffic. One recent result correlates the instability of equilibria in a multiagent system with an absence of gridlocks. These control barrier function–based algorithms do not include a decision-making component—the action is continuous, and negotiation happens through instability. This mechanism, referred to as instinctive negotiation, is contrasted with discontinuity-induced decisions arising from nonconvex optimization. Based on observed behavioral similarities and insights into human implicit and explicit learning, this article proposes a connection with human driving and suggests that humans may employ a mechanism similar to instinctive negotiation to navigate dense traffic. Expected final online publication date for the Annual Review of Control, Robotics, and Autonomous Systems, Volume 7 is May 2024. Please see http://www.annualreviews.org/page/journal/pubdates for revised estimates.","PeriodicalId":29961,"journal":{"name":"Annual Review of Control Robotics and Autonomous Systems","volume":"9 2","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-11-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134956849","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-11-14DOI: 10.1146/annurev-control-070523-115155
Nicolás Faedo, John V. Ringwood
The control of wave energy converters (WECs) to maximize power capture is a challenging problem. In particular, the nature of the wave excitation, which is in general panchromatic (or multi-sinusoidal), presents a reciprocating energy source that needs to be rectified through some means. In addition, the development of suitable control-oriented models is also challenging, requiring correct representation of system hydrodynamics and power take-off (PTO) components, while also lending themselves to control synthesis and real-time computational performance, along with a challenging optimal control problem. This article presents a moment-based mathematical framework for the formulation and solution of WEC control. It shows that moments are ideally suited to WEC control in terms of their ability to accurately characterize the nature of the wave excitation force (and the consequent evolutions in the system variables) while also gracefully including hydrodynamic and PTO nonlinearities as well as a natural extension to WEC arrays. Model reduction, to mold the system model into a control-friendly form, is also a feature of this framework. Expected final online publication date for the Annual Review of Control, Robotics, and Autonomous Systems, Volume 7 is May 2024. Please see http://www.annualreviews.org/page/journal/pubdates for revised estimates.
{"title":"A Control Framework for Ocean Wave Energy Conversion Systems: The Potential of Moments","authors":"Nicolás Faedo, John V. Ringwood","doi":"10.1146/annurev-control-070523-115155","DOIUrl":"https://doi.org/10.1146/annurev-control-070523-115155","url":null,"abstract":"The control of wave energy converters (WECs) to maximize power capture is a challenging problem. In particular, the nature of the wave excitation, which is in general panchromatic (or multi-sinusoidal), presents a reciprocating energy source that needs to be rectified through some means. In addition, the development of suitable control-oriented models is also challenging, requiring correct representation of system hydrodynamics and power take-off (PTO) components, while also lending themselves to control synthesis and real-time computational performance, along with a challenging optimal control problem. This article presents a moment-based mathematical framework for the formulation and solution of WEC control. It shows that moments are ideally suited to WEC control in terms of their ability to accurately characterize the nature of the wave excitation force (and the consequent evolutions in the system variables) while also gracefully including hydrodynamic and PTO nonlinearities as well as a natural extension to WEC arrays. Model reduction, to mold the system model into a control-friendly form, is also a feature of this framework. Expected final online publication date for the Annual Review of Control, Robotics, and Autonomous Systems, Volume 7 is May 2024. Please see http://www.annualreviews.org/page/journal/pubdates for revised estimates.","PeriodicalId":29961,"journal":{"name":"Annual Review of Control Robotics and Autonomous Systems","volume":"12 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-11-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134957156","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-11-14DOI: 10.1146/annurev-control-061423-101708
Lucy Y. Pao, Manuel Pusch, Daniel S. Zalkind
Wind energy is recognized worldwide as cost-effective and environmentally friendly, and it is among the fastest-growing sources of electrical energy. To further decrease the cost of wind energy, wind turbines are being designed at ever-larger scales. To expand the deployment of wind energy, wind turbines are also being designed on floating platforms for placement in deep-water locations offshore. Both larger-scale and floating wind turbines pose challenges because of their greater structural loads and deflections. Complex, large-scale systems such as modern wind turbines increasingly require a control co-design approach, whereby the system design and control design are performed in a more integrated fashion. This article reviews recent developments in control co-design of wind turbines. We provide an overview of wind turbine design objectives and constraints, issues in the design of key wind turbine components, modeling of the wind turbine and environment, and controller coupling issues. Wind turbine control functions and the integration of control design in co-design are detailed with a focus on co-design compatible control approaches. Expected final online publication date for the Annual Review of Control, Robotics, and Autonomous Systems, Volume 7 is May 2024. Please see http://www.annualreviews.org/page/journal/pubdates for revised estimates.
{"title":"Control Co-Design of Wind Turbines","authors":"Lucy Y. Pao, Manuel Pusch, Daniel S. Zalkind","doi":"10.1146/annurev-control-061423-101708","DOIUrl":"https://doi.org/10.1146/annurev-control-061423-101708","url":null,"abstract":"Wind energy is recognized worldwide as cost-effective and environmentally friendly, and it is among the fastest-growing sources of electrical energy. To further decrease the cost of wind energy, wind turbines are being designed at ever-larger scales. To expand the deployment of wind energy, wind turbines are also being designed on floating platforms for placement in deep-water locations offshore. Both larger-scale and floating wind turbines pose challenges because of their greater structural loads and deflections. Complex, large-scale systems such as modern wind turbines increasingly require a control co-design approach, whereby the system design and control design are performed in a more integrated fashion. This article reviews recent developments in control co-design of wind turbines. We provide an overview of wind turbine design objectives and constraints, issues in the design of key wind turbine components, modeling of the wind turbine and environment, and controller coupling issues. Wind turbine control functions and the integration of control design in co-design are detailed with a focus on co-design compatible control approaches. Expected final online publication date for the Annual Review of Control, Robotics, and Autonomous Systems, Volume 7 is May 2024. Please see http://www.annualreviews.org/page/journal/pubdates for revised estimates.","PeriodicalId":29961,"journal":{"name":"Annual Review of Control Robotics and Autonomous Systems","volume":"6 3","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-11-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134956675","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-11-06DOI: 10.1146/annurev-control-062323-102456
Steven M. LaValle, Evan G. Center, Timo Ojala, Matti Pouke, Nicoletta Prencipe, Basak Sakcak, Markku Suomalainen, Kalle G. Timperi, Vadim Weinstein
This article makes the case that a powerful new discipline, which we term perception engineering, is steadily emerging. It follows from a progression of ideas that involve creating illusions, from historical paintings and film to modern video games and virtual reality. Rather than creating physical artifacts such as bridges, airplanes, or computers, perception engineers create illusory perceptual experiences. The scope is defined over any agent that interacts with the physical world, including both biological organisms (humans and animals) and engineered systems (robots and autonomous systems). The key idea is that an agent, called a producer, alters the environment with the intent to alter the perceptual experience of another agent, called a receiver. Most importantly, the article introduces a precise mathematical formulation of this process, based on the von Neumann–Morgenstern notion of information, to help scope and define the discipline. This formulation is then applied to the cases of engineered and biological agents, with discussion of its implications for existing fields such as virtual reality, robotics, and even social media. Finally, open challenges and opportunities for involvement are identified. Expected final online publication date for the Annual Review of Control, Robotics, and Autonomous Systems, Volume 7 is May 2024. Please see http://www.annualreviews.org/page/journal/pubdates for revised estimates.
{"title":"From Virtual Reality to the Emerging Discipline of Perception Engineering","authors":"Steven M. LaValle, Evan G. Center, Timo Ojala, Matti Pouke, Nicoletta Prencipe, Basak Sakcak, Markku Suomalainen, Kalle G. Timperi, Vadim Weinstein","doi":"10.1146/annurev-control-062323-102456","DOIUrl":"https://doi.org/10.1146/annurev-control-062323-102456","url":null,"abstract":"This article makes the case that a powerful new discipline, which we term perception engineering, is steadily emerging. It follows from a progression of ideas that involve creating illusions, from historical paintings and film to modern video games and virtual reality. Rather than creating physical artifacts such as bridges, airplanes, or computers, perception engineers create illusory perceptual experiences. The scope is defined over any agent that interacts with the physical world, including both biological organisms (humans and animals) and engineered systems (robots and autonomous systems). The key idea is that an agent, called a producer, alters the environment with the intent to alter the perceptual experience of another agent, called a receiver. Most importantly, the article introduces a precise mathematical formulation of this process, based on the von Neumann–Morgenstern notion of information, to help scope and define the discipline. This formulation is then applied to the cases of engineered and biological agents, with discussion of its implications for existing fields such as virtual reality, robotics, and even social media. Finally, open challenges and opportunities for involvement are identified. Expected final online publication date for the Annual Review of Control, Robotics, and Autonomous Systems, Volume 7 is May 2024. Please see http://www.annualreviews.org/page/journal/pubdates for revised estimates.","PeriodicalId":29961,"journal":{"name":"Annual Review of Control Robotics and Autonomous Systems","volume":"13 3","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-11-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135589113","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-10-23DOI: 10.1146/annurev-control-062023-082238
Carme Torras
Focus on the ethics of a given technology tends to lag far behind its development. This lag has been particularly acute in the case of artificial intelligence, whose accelerated deployment in a wide range of domains has triggered unprecedented attention on the risks and consequences for society at large, leading to a myriad of ethics regulations, which are difficult to coordinate and integrate due to their late appearance. The very nature of social robots forces their deployment to occur at a much slower pace, providing an opportunity for a profound reflection on ethics, which is already happening in multidisciplinary teams. This article provides a personal view of the ethics landscape, centered on the particularities of social robotics, with the main issues being ordered along two axes (individual and societal) and grouped into eight categories (human dignity, human autonomy, robot transparency, emotional bonding, privacy and safety, justice, freedom, and responsibility). This structure stems from the experience of developing and teaching a university course on ethics in social robotics, whose pedagogical materials are freely available. Expected final online publication date for the Annual Review of Control, Robotics, and Autonomous Systems, Volume 7 is May 2024. Please see http://www.annualreviews.org/page/journal/pubdates for revised estimates.
{"title":"Ethics of Social Robotics: Individual and Societal Concerns and Opportunities","authors":"Carme Torras","doi":"10.1146/annurev-control-062023-082238","DOIUrl":"https://doi.org/10.1146/annurev-control-062023-082238","url":null,"abstract":"Focus on the ethics of a given technology tends to lag far behind its development. This lag has been particularly acute in the case of artificial intelligence, whose accelerated deployment in a wide range of domains has triggered unprecedented attention on the risks and consequences for society at large, leading to a myriad of ethics regulations, which are difficult to coordinate and integrate due to their late appearance. The very nature of social robots forces their deployment to occur at a much slower pace, providing an opportunity for a profound reflection on ethics, which is already happening in multidisciplinary teams. This article provides a personal view of the ethics landscape, centered on the particularities of social robotics, with the main issues being ordered along two axes (individual and societal) and grouped into eight categories (human dignity, human autonomy, robot transparency, emotional bonding, privacy and safety, justice, freedom, and responsibility). This structure stems from the experience of developing and teaching a university course on ethics in social robotics, whose pedagogical materials are freely available. Expected final online publication date for the Annual Review of Control, Robotics, and Autonomous Systems, Volume 7 is May 2024. Please see http://www.annualreviews.org/page/journal/pubdates for revised estimates.","PeriodicalId":29961,"journal":{"name":"Annual Review of Control Robotics and Autonomous Systems","volume":"41 6","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-10-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135366968","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-10-20DOI: 10.1146/annurev-control-061323-095841
Jesse Haviland, Peter Corke
Robotics is powered by software. Software tools control the rate of innovation in robotics research, drive the growth of the robotics industry, and power the education of future innovators and developers. Nearly 900,000 open-source repositories on GitHub are tagged with the keyword robotics—a potentially vast resource, but only a fraction of those are truly accessible in terms of quality, licensability, understandability, and total cost of ownership. The challenge is to match this resource to the needs of students, researchers, and companies to power cutting-edge research and real-world industrial solutions. This article reviews software tools for robotics, including both those created by the community at large and those created by the authors, as well as their impact on education, research, and industry. Expected final online publication date for the Annual Review of Control, Robotics, and Autonomous Systems, Volume 7 is May 2024. Please see http://www.annualreviews.org/page/journal/pubdates for revised estimates.
{"title":"Robotics Software: Past, Present, and Future","authors":"Jesse Haviland, Peter Corke","doi":"10.1146/annurev-control-061323-095841","DOIUrl":"https://doi.org/10.1146/annurev-control-061323-095841","url":null,"abstract":"Robotics is powered by software. Software tools control the rate of innovation in robotics research, drive the growth of the robotics industry, and power the education of future innovators and developers. Nearly 900,000 open-source repositories on GitHub are tagged with the keyword robotics—a potentially vast resource, but only a fraction of those are truly accessible in terms of quality, licensability, understandability, and total cost of ownership. The challenge is to match this resource to the needs of students, researchers, and companies to power cutting-edge research and real-world industrial solutions. This article reviews software tools for robotics, including both those created by the community at large and those created by the authors, as well as their impact on education, research, and industry. Expected final online publication date for the Annual Review of Control, Robotics, and Autonomous Systems, Volume 7 is May 2024. Please see http://www.annualreviews.org/page/journal/pubdates for revised estimates.","PeriodicalId":29961,"journal":{"name":"Annual Review of Control Robotics and Autonomous Systems","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-10-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135567913","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-10-20DOI: 10.1146/annurev-control-061523-102310
Taeyoon Lee, Jaewoon Kwon, Patrick M. Wensing, Frank C. Park
In recent years, the increasing complexity and safety-critical nature of robotic tasks have highlighted the importance of accurate and reliable robot models. This trend has led to a growing belief that, given enough data, traditional physics-based robot models can be replaced by appropriately trained deep networks or their variants. Simultaneously, there has been a renewed interest in physics-based simulation, fueled by the widespread use of simulators to train reinforcement learning algorithms in the sim-to-real paradigm. The primary objective of this review is to present a unified perspective on the process of determining robot models from data, commonly known as system identification or model learning in different subfields. The review aims to illuminate the key challenges encountered and highlight recent advancements in system identification for robotics. Specifically, we focus on recent breakthroughs that leverage the geometry of the identification problem and incorporate physics-based knowledge beyond mere first-principles model parameterizations. Through these efforts, we strive to provide a contemporary outlook on this problem, bridging classical findings with the latest progress in the field. Expected final online publication date for the Annual Review of Control, Robotics, and Autonomous Systems, Volume 7 is May 2024. Please see http://www.annualreviews.org/page/journal/pubdates for revised estimates.
{"title":"Robot Model Identification and Learning: A Modern Perspective","authors":"Taeyoon Lee, Jaewoon Kwon, Patrick M. Wensing, Frank C. Park","doi":"10.1146/annurev-control-061523-102310","DOIUrl":"https://doi.org/10.1146/annurev-control-061523-102310","url":null,"abstract":"In recent years, the increasing complexity and safety-critical nature of robotic tasks have highlighted the importance of accurate and reliable robot models. This trend has led to a growing belief that, given enough data, traditional physics-based robot models can be replaced by appropriately trained deep networks or their variants. Simultaneously, there has been a renewed interest in physics-based simulation, fueled by the widespread use of simulators to train reinforcement learning algorithms in the sim-to-real paradigm. The primary objective of this review is to present a unified perspective on the process of determining robot models from data, commonly known as system identification or model learning in different subfields. The review aims to illuminate the key challenges encountered and highlight recent advancements in system identification for robotics. Specifically, we focus on recent breakthroughs that leverage the geometry of the identification problem and incorporate physics-based knowledge beyond mere first-principles model parameterizations. Through these efforts, we strive to provide a contemporary outlook on this problem, bridging classical findings with the latest progress in the field. Expected final online publication date for the Annual Review of Control, Robotics, and Autonomous Systems, Volume 7 is May 2024. Please see http://www.annualreviews.org/page/journal/pubdates for revised estimates.","PeriodicalId":29961,"journal":{"name":"Annual Review of Control Robotics and Autonomous Systems","volume":"37 11","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-10-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135568032","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-05-03DOI: 10.1146/annurev-control-042920-020021
Bin Hu, Kaiqing Zhang, Na Li, Mehran Mesbahi, Maryam Fazel, Tamer Başar
Gradient-based methods have been widely used for system design and optimization in diverse application domains. Recently, there has been a renewed interest in studying theoretical properties of these methods in the context of control and reinforcement learning. This article surveys some of the recent developments on policy optimization, a gradient-based iterative approach for feedback control synthesis that has been popularized by successes of reinforcement learning. We take an interdisciplinary perspective in our exposition that connects control theory, reinforcement learning, and large-scale optimization. We review a number of recently developed theoretical results on the optimization landscape, global convergence, and sample complexityof gradient-based methods for various continuous control problems, such as the linear quadratic regulator (LQR), [Formula: see text] control, risk-sensitive control, linear quadratic Gaussian (LQG) control, and output feedback synthesis. In conjunction with these optimization results, we also discuss how direct policy optimization handles stability and robustness concerns in learning-based control, two main desiderata in control engineering. We conclude the survey by pointing out several challenges and opportunities at the intersection of learning and control.
{"title":"Toward a Theoretical Foundation of Policy Optimization for Learning Control Policies","authors":"Bin Hu, Kaiqing Zhang, Na Li, Mehran Mesbahi, Maryam Fazel, Tamer Başar","doi":"10.1146/annurev-control-042920-020021","DOIUrl":"https://doi.org/10.1146/annurev-control-042920-020021","url":null,"abstract":"Gradient-based methods have been widely used for system design and optimization in diverse application domains. Recently, there has been a renewed interest in studying theoretical properties of these methods in the context of control and reinforcement learning. This article surveys some of the recent developments on policy optimization, a gradient-based iterative approach for feedback control synthesis that has been popularized by successes of reinforcement learning. We take an interdisciplinary perspective in our exposition that connects control theory, reinforcement learning, and large-scale optimization. We review a number of recently developed theoretical results on the optimization landscape, global convergence, and sample complexityof gradient-based methods for various continuous control problems, such as the linear quadratic regulator (LQR), [Formula: see text] control, risk-sensitive control, linear quadratic Gaussian (LQG) control, and output feedback synthesis. In conjunction with these optimization results, we also discuss how direct policy optimization handles stability and robustness concerns in learning-based control, two main desiderata in control engineering. We conclude the survey by pointing out several challenges and opportunities at the intersection of learning and control.","PeriodicalId":29961,"journal":{"name":"Annual Review of Control Robotics and Autonomous Systems","volume":"15 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-05-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134923116","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：