Learning to Navigate Endoscopic Capsule Robots

Turan, Mehmet; Almalioglu, Yasin; Gilbert, Hunter B.; Mahmood, Faisal; Durr, Nicholas J.; Araújo, Hélder; Sari, Alp Eren; Ajay, Anurag; Sitti, Metin

doi:10.1109/lra.2019.2924846

Cited by 21 publications

(10 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For magnetic capsule robots or magnet-tipped tethered devices for medical applications, whose size can range from a few millimeters to centimeters in terms of the device diameter, localization and tracking of the device have been obtained from internal or external sensing (e.g., embedded Hall-effect sensors and inertial measurement units ,, or externally arranged Hall-effect sensor arrays ,,− ) or real-time imaging (e.g., internal imaging using endoscopy cameras, , external imaging through optical or stereocameras, ,,, X-ray fluoroscopy, , magnetic resonance imaging, or ultrasound). The state observation from device tracking and localization through real-time sensing or visual feedback through imaging allows for the implementation of closed-loop feedback control or learning-based, data-driven control for intelligent magnetic manipulation. For small-scale magnetic soft robots, however, device tracking and localization for closed-loop control and advanced manipulation through internal or external sensing can be challenging due to their constrained size, and the state observation would rely mostly on the visual feedback from real-time external imaging.…”

Section: Considerations For Future Developmentsmentioning

confidence: 99%

Magnetic Soft Materials and Robots

2022

View full text Add to dashboard Cite

In conventional classification, soft robots feature mechanical compliance as the main distinguishing factor from traditional robots made of rigid materials. Recent advances in functional soft materials have facilitated the emergence of a new class of soft robots capable of tether-free actuation in response to external stimuli such as heat, light, solvent, or electric or magnetic field. Among the various types of stimuli-responsive materials, magnetic soft materials have shown remarkable progress in their design and fabrication, leading to the development of magnetic soft robots with unique advantages and potential for many important applications. However, the field of magnetic soft robots is still in its infancy and requires further advancements in terms of design principles, fabrication methods, control mechanisms, and sensing modalities. Successful future development of magnetic soft robots would require a comprehensive understanding of the fundamental principle of magnetic actuation, as well as the physical properties and behavior of magnetic soft materials. In this review, we discuss recent progress in the design and fabrication, modeling and simulation, and actuation and control of magnetic soft materials and robots. We then give a set of design guidelines for optimal actuation performance of magnetic soft materials. Lastly, we summarize potential biomedical applications of magnetic soft robots and provide our perspectives on next-generation magnetic soft robots.

show abstract

Section: Considerations For Future Developmentsmentioning

confidence: 99%

Magnetic Soft Materials and Robots

2022

View full text Add to dashboard Cite

show abstract

“…Ref. [100], and omits the contribution of the electromagnet's magnetic field from the measured magnetic field, and calculates the 5-DOF information of the capsule in real-time using a nonlinear optimization algorithm. Thus, with this system, accurate inspection of the GI tract has been achieved through active control of CEs, multi-sensor positioning systems, and combined deep learning and sensor fusion techniques [100] .…”

Section: Electromagnetic Massmentioning

confidence: 99%

Magnetic Actuation Systems and Magnetic Robots for Gastrointestinal Examination and Treatment

Sun

Liu

Wang

2023

Chin. J. Electr. Eng.

View full text Add to dashboard Cite

Magnetic actuation technology (MAT) provides novel diagnostic tools for the early screening and treatment of digestive cancers, which have high morbidity and mortality rates worldwide. The application of magnetic actuation systems and magnetic robots in gastrointestinal (GI) diagnosis and treatment to provide a comprehensive reference manual for scholars in the field of MAT research are reviewed. It describes the basic principles of magnetic actuation and magnetic field safety, introduces the design, manufacturing, control, and performance parameters of magnetic actuation systems, as well as the applicability and limitations of each system for different parts of the GI tract. It analyzes the characteristics and advantages of different types and functions of magnetic robots, summarizes the challenges faced by MAT in clinical applications, and provides an outlook on the future prospects of the field.

show abstract

“…[48] In other words, SAC learns a policy that successfully completes the task while acting as randomly as possible, which in practice often leads to robust policies that are tolerant of perturbations in environmental conditions. [37] SAC had previously proven useful for real-world robotic tasks with high-dimensional, continuous state and action spaces, [37,39] which suggested that it would be applicable to our microrobotic control problem. In previously reported applications of real-world RL with physical systems, [60] SAC was demonstrated to be highly sample-efficient, requiring relatively few environmental interactions in order to develop a successful policy.…”

Section: Entropy-regularized Deep Rl Enabled Continuous Microrobot Co...mentioning

confidence: 99%

“…RL algorithms have achieved success in a range of complex robotic control applications. [36][37][38][39][40][41][42] For example, RL for robotic control has been demonstrated to create robotic control policies that achieve better performance than many humans at complex tasks such as grasping and accurate throwing of irregularly shaped objects into bins. [40] RL algorithms have also been shown to exceed human-level performance in complex virtual tasks with large possible state spaces that cannot be tractably and exhaustively modeled, such as the game of Go.…”

mentioning

confidence: 99%

Smart Magnetic Microrobots Learn to Swim with Deep Reinforcement Learning

Behrens

Ruder

2022

Advanced Intelligent Systems

View full text Add to dashboard Cite

Swimming microrobots are increasingly developed with complex materials and dynamic shapes and are expected to operate in complex environments in which the system dynamics are difficult to model and positional control of the microrobot is not straightforward to achieve. Deep reinforcement learning is a promising method of autonomously developing robust controllers for creating smart microrobots, which can adapt their behavior to operate in uncharacterized environments without the need to model the system dynamics. This article reports the development of a smart helical magnetic hydrogel microrobot that uses the soft actor critic reinforcement learning algorithm to autonomously derive a control policy which allows the microrobot to swim through an uncharacterized biomimetic fluidic environment under control of a time‐varying magnetic field generated from a three‐axis array of electromagnets. The reinforcement learning agent learns successful control policies from both state vector input and raw images, and the control policies learned by the agent recapitulate the behavior of rationally designed controllers based on physical models of helical swimming microrobots. Deep reinforcement learning applied to microrobot control is likely to significantly expand the capabilities of the next generation of microrobots.

show abstract

Learning to Navigate Endoscopic Capsule Robots

Cited by 21 publications

References 18 publications

Magnetic Soft Materials and Robots

Magnetic Soft Materials and Robots

Magnetic Actuation Systems and Magnetic Robots for Gastrointestinal Examination and Treatment

Smart Magnetic Microrobots Learn to Swim with Deep Reinforcement Learning

Contact Info

Product

Resources

About