Visual Case Retrieval for Interpreting Skill Demonstrations

Fitzgerald, Tesca; McGreggor, Keith; Akgün, Barış; Thomaz, Andrea L.; Goel, Ashok K.

doi:10.1007/978-3-319-24586-7_9

Cited by 8 publications

(5 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Despite the challenges, learning from passive observation has been successfully applied to various tasks, such as collaborative furniture assembly (39), autonomous driving (40), table-top actions (41,42), and knot tying (43). In some cases, the human user is not observed directly, and only the objects in the scene are tracked (31,44).…”

Section: Passive Observationmentioning

confidence: 99%

Recent Advances in Robot Learning from Demonstration

Ravichandar

Polydoros

Chernova

et al. 2020

Annu. Rev. Control Robot. Auton. Syst.

501

219

View full text Add to dashboard Cite

In the context of robotics and automation, learning from demonstration (LfD) is the paradigm in which robots acquire new skills by learning to imitate an expert. The choice of LfD over other robot learning methods is compelling when ideal behavior can be neither easily scripted (as is done in traditional robot programming) nor easily defined as an optimization problem, but can be demonstrated. While there have been multiple surveys of this field in the past, there is a need for a new one given the considerable growth in the number of publications in recent years. This review aims to provide an overview of the collection of machine-learning methods used to enable a robot to learn from and imitate a teacher. We focus on recent advancements in the field and present an updated taxonomy and characterization of existing methods. We also discuss mature and emerging application areas for LfD and highlight the significant challenges that remain to be overcome both in theory and in practice.

show abstract

Section: Passive Observationmentioning

confidence: 99%

Recent Advances in Robot Learning from Demonstration

Ravichandar

Polydoros

Chernova

et al. 2020

Annu. Rev. Control Robot. Auton. Syst.

501

219

View full text Add to dashboard Cite

show abstract

“…Many studies on robotic demonstration learning focus on mapping and replicating human movements, with the goal of maximizing the imitation of precise motion trajectories associated with the set task [7][8][9][10][11][12]. For example, Fitzgerald et al [13] used "kinesthetic teaching" to complete the QR code block rearrangement task, and Wu et al [14] marked benchmark points on objects and the demonstrator's arms and used visual motion tracking methods to record the trajectory of the benchmark points.…”

Section: Related Workmentioning

confidence: 99%

Research on LFD System of Humanoid Dual-Arm Robot

Cui,

Kou,

Chen

et al. 2024

Symmetry

View full text Add to dashboard Cite

Although robots have been widely used in a variety of fields, the idea of enabling them to perform multiple tasks in the same way that humans do remains a difficulty. To solve this, we investigate the learning from demonstration (LFD) system with our independently designed symmetrical humanoid dual-arm robot. We present a novel action feature matching algorithm. This algorithm accurately transforms human demonstration data into task models that robots can directly execute, considerably improving LFD’s generalization capabilities. In our studies, we used motion capture cameras to capture human demonstration actions, which included combinations of simple actions (the action layer) and a succession of complicated operational tasks (the task layer). For the action layer data, we employed Gaussian mixture models (GMM) for processing and constructing an action primitive library. As for the task layer data, we created a “keyframe” segmentation method to transform this data into a series of action primitives and build another action primitive library. Guided by our algorithm, the robot successfully imitated complex human tasks. Results show its excellent task learning and execution, providing an effective solution for robots to learn from human demonstrations and significantly advancing robot technology.

show abstract

“…El RBC es un método de inteligencia artificial para solucionar problemas no estructurados, donde el razonamiento se realiza a partir de una memoria asociativa, que usa un algoritmo para determinar una medida de semejanza entre dos objetos. En los sistemas basados en casos el dominio no tiene que estar representado por completo [39] y el aprendizaje es progresivo [40].…”

Section: El Razonamiento Basado En Casos Con El Uso De La Ontologíaunclassified

Sistema Inteligente de Información Geográfica para las empresas eléctricas cubanas

Fleitas

Rodríguez

Lorenzo

2019

Ingeniare. Rev. chil. ing.

View full text Add to dashboard Cite

RESUMENLas empresas eléctricas necesitan métodos eficaces de visualizar información. En la Unión Eléctrica en Cuba se trabaja en el desarrollo de un Sistema de Información Geográfica (SIG) que cuente con una base conceptual y dé respuesta a las distintas peticiones del usuario. Para ello, el objetivo de la investigación es desarrollar un modelo para el manejo de datos geoespaciales, con el uso de técnicas de Inteligencia Artificial, como soporte a la toma de decisiones en el sector eléctrico cubano. Como primer paso se elabora una ontología ligera que dote al sistema de la base conceptual. Para lograr las consultas automáticas se desarrolla un Sistema de Razonamiento Basado en Casos. La base de casos contiene la descripción de consultas estáticas realizadas previamente en forma de casos. Cada consulta se compone por once rasgos fundamentales derivados de la ingeniería del conocimiento realizada, de los cuales ocho son predictores y tres objetivos. La calidad final del SIG es verificada de acuerdo a los estándares de calidad de la norma ISO 9126:2002. Este sistema se aplica en todas las empresas eléctricas del país en las distintas áreas que inciden en los procesos de transmisión. Los resultados de la validación en campo evidencian la factibilidad de la propuesta.Palabras clave: Sistema de información geográfica, sistemas basados en el conocimiento, ontologías. ABSTRACTElectric companies need effective methods to visualize information. The Electric Union in Cuba works on the development of a Geographic Information System (GIS) with a conceptual basis that responds to the different user requests. For this, the research objective is to develop a model for the management of geospatial data, with the use of artificial intelligence techniques, as support for decision making in the Cuban electric sector. As a first step, a lightweight ontology that endows a conceptual basis is elaborated. To achieve automatic queries, a Case-Based Reasoning System is developed. The case database contain the description of static queries previously made in the form of cases. Each query is made up of eleven fundamentals traits derived of the completed knowledge engineering performed, of which: eight are predictors and three are objectives. The final quality of the GIS is verified according to the quality standards of ISO 9126:2002 norm. The system is applied in all electric companies of the country in the different areas that affect the transmission processes. The results of field validation evidence the feasibility of the proposal.

show abstract

Visual Case Retrieval for Interpreting Skill Demonstrations

Cited by 8 publications

References 20 publications

Recent Advances in Robot Learning from Demonstration

Recent Advances in Robot Learning from Demonstration

Research on LFD System of Humanoid Dual-Arm Robot

Sistema Inteligente de Información Geográfica para las empresas eléctricas cubanas

Contact Info

Product

Resources

About