Deep learning of biomimetic sensorimotor control for biomechanical human animation

Nakada, Masaki; Zhou, Tao; Chen, Honglin; Weiss, Tomer; Terzopoulos, Demetri

doi:10.1145/3197517.3201305

Cited by 47 publications

(37 citation statements)

References 29 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Lee et al 2014]. Recently, Nakada et al [2018] introduced a sensorimotor system that directly maps the photoreceptor responses to muscle activations, bestowing a visuomotor controller of the character's eyes, head, and limbs for non-trivial, visually-guided tasks. Of particular interest to us is the use of training data synthesized by a muscle-based simulator [S.H.…”

Section: Related Workmentioning

confidence: 99%

Synthesis of biologically realistic human motion using joint torque actuation

et al. 2019

View full text Add to dashboard Cite

Using joint actuators to drive the skeletal movements is a common practice in character animation, but the resultant torque patterns are often unnatural or infeasible for real humans to achieve. On the other hand, physiologicallybased models explicitly simulate muscles and tendons and thus produce more human-like movements and torque patterns. This paper introduces a technique to transform an optimal control problem formulated in the muscleactuation space to an equivalent problem in the joint-actuation space, such that the solutions to both problems have the same optimal value. By solving the equivalent problem in the joint-actuation space, we can generate humanlike motions comparable to those generated by musculotendon models, while retaining the benefit of simple modeling and fast computation offered by joint-actuation models. Our method transforms constant bounds on muscle activations to nonlinear, state-dependent torque limits in the joint-actuation space. In addition, the metabolic energy function on muscle activations is transformed to a nonlinear function of joint torques, joint configuration and joint velocity. Our technique can also benefit policy optimization using deep reinforcement learning approach, by providing a more anatomically realistic action space for the agent to explore during the learning process. We take the advantage of the physiologically-based simulator, OpenSim, to provide training data for learning the torque limits and the metabolic energy function. Once trained, the same torque limits and the energy function can be applied to drastically different motor tasks formulated as either trajectory optimization or policy learning.

show abstract

Section: Related Workmentioning

confidence: 99%

Synthesis of biologically realistic human motion using joint torque actuation

et al. 2019

View full text Add to dashboard Cite

show abstract

“…The task can be something like the World Chase Tag competition, where two athletes take turns to tag the opponent, using athletic movements, in an arena filled with obstacles [146]. To this end, we plan to develop a simulation environment with a faster physics engine that can handle multi-body contacts between human models and obstacles [3,4,5], provide a human musculoskeletal model with an articulated upper body [88,11], and design a visuomotor control interface [147].…”

Section: Future Directionsmentioning

confidence: 99%

Deep reinforcement learning for modeling human locomotion control in neuromechanical simulation

Song

Kidziński

Peng

et al. 2020

Preprint

View full text Add to dashboard Cite

Modeling human motor control and predicting how humans will move in novel environments is a grand scientific challenge. Despite advances in neuroscience techniques, it is still difficult to measure and interpret the activity of the millions of neurons involved in motor control. Thus, researchers in the fields of biomechanics and motor control have proposed and evaluated motor control models via neuromechanical simulations, which produce physically correct motions of a musculoskeletal model. Typically, researchers have developed control models that encode physiologically plausible motor control hypotheses and compared the resulting simulation behaviors to measurable human motion data. While such plausible control models were able to simulate and explain many basic locomotion behaviors (e.g. walking, running, and climbing stairs), modeling higher layer controls (e.g. processing environment cues, planning long-term motion strategies, and coordinating basic motor skills to navigate in dynamic and complex environments) remains a challenge. Recent advances in deep reinforcement learning lay a foundation for modeling these complex control processes and controlling a diverse repertoire of human movement; however, reinforcement learning has been rarely applied in neuromechanical simulation to model human control. In this paper, we review the current state of neuromechanical simulations, along with the fundamentals of reinforcement learning, as it applies to human locomotion. We also present a scientific competition and accompanying software platform, which we have organized to accelerate the use of reinforcement learning in neuromechanical simulations. This “Learn to Move” competition, which we have run annually since 2017 at the NeurIPS conference, has attracted over 1300 teams from around the world. Top teams adapted state-of-art deep reinforcement learning techniques to produce complex motions, such as quick turning and walk-to-stand transitions, that have not been demonstrated before in neuromechanical simulations without utilizing reference motion data. We close with a discussion of future opportunities at the intersection of human movement simulation and reinforcement learning and our plans to extend the Learn to Move competition to further facilitate interdisciplinary collaboration in modeling human motor control for biomechanics and rehabilitation research.

show abstract

“…Our approach to eye modeling and simulation is inspired mainly by our recent work [Nakada et al 2018b] that introduced an elaborate, biomimetic sensorimotor system for a full-body, biomechanical human musculoskeletal model whose neural control mechanisms are based on deep learning. Unlike the uniform, Cartesian grid arrangement of most artificial imaging sensors, visual sampling in the primate retina is known to be strongly nonuniform [Schwartz 1977].…”

Section: Related Workmentioning

confidence: 99%

“…Unlike the uniform, Cartesian grid arrangement of most artificial imaging sensors, visual sampling in the primate retina is known to be strongly nonuniform [Schwartz 1977]. Accordingly, our human model in [Nakada et al 2018b] had biologically inspired eyes with foveated retinas (see also [Nakada et al 2018a]). However, they were overly simplistic, purely kinematic eyes modeled as pinhole cameras.…”

Section: Related Workmentioning

confidence: 99%

“…There are approximately 5 million cones in the human retina [Deering 2005]. With this many photoreceptors in its ONV input, training a fully-connected neural network-as we did in [Nakada et al 2018b] with only 3,600 retinal photoreceptors-is intractable due to the overwhelming memory requirements. Furthermore, the popular Convolutional Neural Network (CNN), which is designed for conventional, regularly-sampled image arrays, is inappropriate for use with our irregular distribution of retinal photoreceptors.…”

Section: Linet Foveation Dnnmentioning

confidence: 99%

See 1 more Smart Citation

Biomimetic eye modeling & deep neuromuscular oculomotor control

et al. 2019

Self Cite

View full text Add to dashboard Cite

We present a novel, biomimetic model of the eye for realistic virtual human animation. We also introduce a deep learning approach to oculomotor control that is compatible with our biomechanical eye model. Our eye model consists of the following functional components: (i) submodels of the 6 extraocular muscles that actuate realistic eye movements, (ii) an iris submodel, actuated by pupillary muscles, that accommodates to incoming light intensity, (iii) a corneal submodel and a deformable, ciliary-muscle-actuated lens submodel, which refract incoming light rays for focal accommodation, and (iv) a retina with a multitude of photoreceptors arranged in a biomimetic, foveated distribution. The light intensity captured by the photoreceptors is computed using ray tracing from the photoreceptor positions through the finite aperture pupil into the 3D virtual environment, and the visual information from the retina is output via an optic nerve vector. Our oculomotor control system includes a foveation controller implemented as a locally-connected, irregular Deep Neural Network (DNN), or "LiNet", that conforms to the nonuniform retinal photoreceptor distribution, and a neuromuscular motor controller implemented as a fully-connected DNN, plus auxiliary Shallow Neural Networks (SNNs) that control the accommodation of the pupil and lens. The DNNs are trained offline through deep learning from data synthesized by the eye model itself. Once trained, the oculomotor control system operates robustly and efficiently online. It innervates the intraocular muscles to perform illumination and focal accommodation and the extraocular muscles to produce natural eye movements in order to foveate and pursue moving visual targets. We additionally demonstrate the operation of our eye model (binocularly) within our recently introduced sensorimotor control framework involving an anatomically-accurate biomechanical human musculoskeletal model.

show abstract

Deep learning of biomimetic sensorimotor control for biomechanical human animation

Cited by 47 publications

References 29 publications

Synthesis of biologically realistic human motion using joint torque actuation

Synthesis of biologically realistic human motion using joint torque actuation

Deep reinforcement learning for modeling human locomotion control in neuromechanical simulation

Biomimetic eye modeling & deep neuromuscular oculomotor control

Contact Info

Product

Resources

About