اشترك بالحزمة الذهبية واحصل على وصول غير محدود شمرا أكاديميا
تسجيل مستخدم جديدNeuromechanics-based Deep Reinforcement Learning of Neurostimulation Control in FES cycling
50
0
0.0
(
0
)
نشر من قبل
Mahendran Subramanian
تاريخ النشر
2021
مجال البحث
الهندسة المعلوماتية
والبحث باللغة
English
اسأل ChatGPT حول البحث
ﻻ يوجد ملخص باللغة العربية
Functional Electrical Stimulation (FES) can restore motion to a paralysed persons muscles. Yet, control stimulating many muscles to restore the practical function of entire limbs is an unsolved problem. Current neurostimulation engineering still relies on 20th Century control approaches and correspondingly shows only modest results that require daily tinkering to operate at all. Here, we present our state of the art Deep Reinforcement Learning (RL) developed for real time adaptive neurostimulation of paralysed legs for FES cycling. Core to our approach is the integration of a personalised neuromechanical component into our reinforcement learning framework that allows us to train the model efficiently without demanding extended training sessions with the patient and working out of the box. Our neuromechanical component includes merges musculoskeletal models of muscle and or tendon function and a multistate model of muscle fatigue, to render the neurostimulation responsive to a paraplegics cyclist instantaneous muscle capacity. Our RL approach outperforms PID and Fuzzy Logic controllers in accuracy and performance. Crucially, our system learned to stimulate a cyclists legs from ramping up speed at the start to maintaining a high cadence in steady state racing as the muscles fatigue. A part of our RL neurostimulation system has been successfully deployed at the Cybathlon 2020 bionic Olympics in the FES discipline with our paraplegic cyclist winning the Silver medal among 9 competing teams.
قيم البحث
اقرأ أيضاً
Recent breakthroughs in Go play and strategic games have witnessed the great potential of reinforcement learning in intelligently scheduling in uncertain environment, but some bottlenecks are also encountered when we generalize this paradigm to unive
rsal complex tasks. Among them, the low efficiency of data utilization in model-free reinforcement algorithms is of great concern. In contrast, the model-based reinforcement learning algorithms can reveal underlying dynamics in learning environments and seldom suffer the data utilization problem. To address the problem, a model-based reinforcement learning algorithm with attention mechanism embedded is proposed as an extension of World Models in this paper. We learn the environment model through Mixture Density Network Recurrent Network(MDN-RNN) for agents to interact, with combinations of variational auto-encoder(VAE) and attention incorporated in state value estimates during the process of learning policy. In this way, agent can learn optimal policies through less interactions with actual environment, and final experiments demonstrate the effectiveness of our model in control problem.
Model-based reinforcement learning (MBRL) is believed to have much higher sample efficiency compared to model-free algorithms by learning a predictive model of the environment. However, the performance of MBRL highly relies on the quality of the lear
ned model, which is usually built in a black-box manner and may have poor predictive accuracy outside of the data distribution. The deficiencies of the learned model may prevent the policy from being fully optimized. Although some uncertainty analysis-based remedies have been proposed to alleviate this issue, model bias still poses a great challenge for MBRL. In this work, we propose to leverage the prior knowledge of underlying physics of the environment, where the governing laws are (partially) known. In particular, we developed a physics-informed MBRL framework, where governing equations and physical constraints are utilized to inform the model learning and policy search. By incorporating the prior information of the environment, the quality of the learned model can be notably improved, while the required interactions with the environment are significantly reduced, leading to better sample efficiency and learning performance. The effectiveness and merit have been demonstrated over a handful of classic control problems, where the environments are governed by canonical ordinary/partial differential equations.
Reinforcement learning algorithms rely on exploration to discover new behaviors, which is typically achieved by following a stochastic policy. In continuous control tasks, policies with a Gaussian distribution have been widely adopted. Gaussian explo
ration however does not result in smooth trajectories that generally correspond to safe and rewarding behaviors in practical tasks. In addition, Gaussian policies do not result in an effective exploration of an environment and become increasingly inefficient as the action rate increases. This contributes to a low sample efficiency often observed in learning continuous control tasks. We introduce a family of stationary autoregressive (AR) stochastic processes to facilitate exploration in continuous control domains. We show that proposed processes possess two desirable features: subsequent process observations are temporally coherent with continuously adjustable degree of coherence, and the process stationary distribution is standard normal. We derive an autoregressive policy (ARP) that implements such processes maintaining the standard agent-environment interface. We show how ARPs can be easily used with the existing off-the-shelf learning algorithms. Empirically we demonstrate that using ARPs results in improved exploration and sample efficiency in both simulated and real world domains, and, furthermore, provides smooth exploration trajectories that enable safe operation of robotic hardware.
Deep reinforcement learning has been recognized as an efficient technique to design optimal strategies for different complex systems without prior knowledge of the control landscape. To achieve a fast and precise control for quantum systems, we propo
se a novel deep reinforcement learning approach by constructing a curriculum consisting of a set of intermediate tasks defined by a fidelity threshold. Tasks among a curriculum can be statically determined using empirical knowledge or adaptively generated with the learning process. By transferring knowledge between two successive tasks and sequencing tasks according to their difficulties, the proposed curriculum-based deep reinforcement learning (CDRL) method enables the agent to focus on easy tasks in the early stage, then move onto difficult tasks, and eventually approaches the final task. Numerical simulations on closed quantum systems and open quantum systems demonstrate that the proposed method exhibits improved control performance for quantum systems and also provides an efficient way to identify optimal strategies with fewer control pulses.
We propose a method for learning expressive energy-based policies for continuous states and actions, which has been feasible only in tabular domains before. We apply our method to learning maximum entropy policies, resulting into a new algorithm, cal
led soft Q-learning, that expresses the optimal policy via a Boltzmann distribution. We use the recently proposed amortized Stein variational gradient descent to learn a stochastic sampling network that approximates samples from this distribution. The benefits of the proposed algorithm include improved exploration and compositionality that allows transferring skills between tasks, which we confirm in simulated experiments with swimming and walking robots. We also draw a connection to actor-critic methods, which can be viewed performing approximate inference on the corresponding energy-based model.
الأسئلة المقترحة
سجل دخول لتتمكن من نشر تعليقات
التعليقات
جاري جلب التعليقات
سجل دخول لتتمكن من متابعة معايير البحث التي قمت باختيارها
