High-Bandwidth Nonlinear Control for Soft Actuators with Recursive Network Models

Manzano, Sarah Aguasvivas; Xu, Patricia; Ly, Khoi; Shepherd, Robert F.; Correll, Nikolaus

doi:10.1007/978-3-030-71151-1_52

Cited by 6 publications

(8 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The physical soft actuator systems studied in this work are: (1) A tendon-based soft actuator with an internal network of optical sensors and a Tensilica Xtensa LX6 (ESP32 Thing, Sparkfun), 10 and (2) A folded HASEL-based, multilayered platform equipped with magnetic sensors and an ARM Cortex-M4f platform (Teensy 3.6, PJRC). 9 The tendon-based platform in Figure 1(a) consists of a soft 3D-printed mesh that has a network of embedded LightLace stretchable sensors, organically weaved inside the soft mesh structure by hand during fabrication.…”

Section: Hardware Testbedsmentioning

confidence: 99%

“…Throughout this work, we use the term "untethered" interchangeably with the term "computationally untethered," i.e., high-performance computation in the firmware of the embedded low-power compute units or units without being tethered to personal computers, clouds, joysticks, or any remote brains and using feedback from proprioception provided by networks of synthetic afferent sensor networks. 9 Aguasvivas Manzano et al 10 presents initial formulations and a Python prototype of the output tracking technique developed in this work for the tendon-based platform with embedded LightLace sensors developed by Xu et al we also study the robustness of this control method against mechanical disturbances. Figure 1(a) displays the actuating, tendon-based, platform that uses feedback from this network of embedded LightLace sensors.…”

Section: Introductionmentioning

confidence: 99%

“…Aguasvivas Manzano et al 10 presents initial formulations and a Python prototype of the output tracking technique developed in this work for the tendon-based platform with embedded LightLace sensors developed by Xu et al we also study the robustness of this control method against mechanical disturbances. Figure 1(a) displays the actuating, tendon-based, platform that uses feedback from this network of embedded LightLace sensors.…”

Section: Introductionmentioning

confidence: 99%

“…Lastly, we test the claim of the tractability of this algorithm in low-power computers by measuring the power and modelling the energy consumption of this algorithm 1 .

Figure 1.Hardware platforms studied for on-board nonlinear neural-based control. (A) A tendon-based platform with embedded LightLace sensors 10 (B) A HASEL-based tilting platform with embedded magnetic sensors. 9 …”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Toward smart composites: Small-scale, untethered prediction and control for soft sensor/actuator systems

Manzano

Sundaram

et al. 2022

Journal of Composite Materials

Self Cite

View full text Add to dashboard Cite

We present formulation and open-source tools to achieve in-material model predictive control of sensor/actuator systems using learned forward kinematics and on-device computation. Microcontroller units that compute the prediction and control task while colocated with the sensors and actuators enable in-material untethered behaviors. In this approach, small parameter size neural network models learn forward kinematics offline. Our open-source compiler, nn4mc, generates code to offload these predictions onto MCUs. A Newton-Raphson solver then computes the control input in real time. We first benchmark this nonlinear control approach against a PID controller on a mass-spring-damper simulation. We then study experimental results on two experimental rigs with different sensing, actuation and computational hardware: a tendon-based platform with embedded LightLace sensors and a HASEL-based platform with magnetic sensors. Experimental results indicate effective high-bandwidth tracking of reference paths (≥120 Hz) with a small memory footprint (≤6.4% of flash memory). The measured path following error does not exceed 2mm in the tendon-based platform. The simulated path following error does not exceed 1mm in the HASEL-based platform. The mean power consumption of this approach in an ARM Cortex-M4f device is 45.4 mW. This control approach is also compatible with Tensorflow Lite models and equivalent on-device code. In-material intelligence enables a new class of composites that infuse autonomy into structures and systems with refined artificial proprioception.

show abstract

Section: Hardware Testbedsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

“…Lastly, we test the claim of the tractability of this algorithm in low-power computers by measuring the power and modelling the energy consumption of this algorithm 1 .

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Toward smart composites: Small-scale, untethered prediction and control for soft sensor/actuator systems

Manzano

Sundaram

et al. 2022

Journal of Composite Materials

Self Cite

View full text Add to dashboard Cite

show abstract

“…employed parsimonious recursive neural networks and Newton‐Raphson algorithms for online optimization of models that provided accurate and updated predictions for a stretchable fiberoptic network providing feedback into a DC motor—tendon based soft actuator system in a feedback loop operating in excess of 100 Hz. [ 135 ]…”

Section: Controlmentioning

confidence: 99%

Elastomeric Haptic Devices for Virtual and Augmented Reality

Bai

Shepherd

2021

Adv Funct Materials

View full text Add to dashboard Cite

Since the modern concepts for virtual and augmented reality are first introduced in the 1960's, the field has strived to develop technologies for immersive user experience in a fully or partially virtual environment. Despite the great progress in visual and auditory technologies, haptics has seen much slower technological advances. The challenge is because skin has densely packed mechanoreceptors distributed over a very large area with complex topography; devising an apparatus as targeted as an audio speaker or television for the localized sensory input of an ear canal or iris is more difficult. Furthermore, the soft and sensitive nature of the skin makes it difficult to apply solid state electronic solutions that can address large areas without causing discomfort. The maturing field of soft robotics offers potential solutions toward this challenge. In this article, the definition and history of virtual (VR) and augmented reality (AR) is first reviewed. Then an overview of haptic output and input technologies is presented, opportunities for soft robotics are identified, and mechanisms of intrinsically soft actuators and sensors are introduced. Finally, soft haptic output and input devices are reviewed with categorization by device forms, and examples of soft haptic devices in VR/AR environments are presented.

show abstract

Embedded Magnetic Sensing for Feedback Control of Soft HASEL Actuators

Sundaram

Johnson

et al. 2023

IEEE Trans. Robot.

View full text Add to dashboard Cite

High-Bandwidth Nonlinear Control for Soft Actuators with Recursive Network Models

Cited by 6 publications

References 10 publications

Toward smart composites: Small-scale, untethered prediction and control for soft sensor/actuator systems

Toward smart composites: Small-scale, untethered prediction and control for soft sensor/actuator systems

Elastomeric Haptic Devices for Virtual and Augmented Reality

Embedded Magnetic Sensing for Feedback Control of Soft HASEL Actuators

Contact Info

Product

Resources

About