Base on -Learning Pareto Optimality for Linear Itô Stochastic Systems With Markovian Jumps<i />

Ming, Zhongyang; Mu, Yunfei; Li, Weihua; Luo, Yanhong

doi:10.1109/tase.2023.3234928

Cited by 9 publications
(4 citation statements)

References 33 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The designed controller (10) is obviously governed by the stochastic signal 𝛼(t), which has been introduced previously. This means the controller (10) does not rely on the system modes r(t), thus showing asynchronous property. Especially, if 𝛼(t) = r(t), the asynchronous controller (10) becomes a synchronous one.…”

Section: Resultsmentioning
confidence: 99%

“…This means the controller (10) does not rely on the system modes r(t), thus showing asynchronous property. Especially, if 𝛼(t) = r(t), the asynchronous controller (10) becomes a synchronous one. In fact, due to the mismatch phenomenon of the sampling time and network dropout, there always exist delays when the controller receives feedback information, such as x(t), from sensors.…”

Section: Resultsmentioning
confidence: 99%

“…Naturally, it is highly desirable to connect MJSs with the Brownian motion, holding great promise for theoretical novelties and practical applications. In recent years, a lot of works have been conducted to yield rich results on MJSs with It $\hat{\mathrm{o}}$ motions [9–24]. In more detail, [10] addressed the stochastic stability problem for semi‐MJSs with It $\hat{\mathrm{o}}$ ‐type nonlinearity.…”

Section: Introductionmentioning
confidence: 99%

“…In recent years, a lot of works have been conducted to yield rich results on MJSs with It $\hat{\mathrm{o}}$ motions [9–24]. In more detail, [10] addressed the stochastic stability problem for semi‐MJSs with It $\hat{\mathrm{o}}$ ‐type nonlinearity. In [14], the global exponential stabilization problem was investigated for delayed stochastic MJSs, in which the Lyapunov‐Krasovskii function was formulated to facilitate analyses.…”

Section: Introductionmentioning
confidence: 99%

See 3 more Smart Citations
Hidden Markov model‐based asynchronous exponential stabilization for stochastic Markovian jump systems with incomplete controller mode information
Lu,
Li,
Tian
2023
Asian Journal of Control
1
0
0
0
View full text Add to dashboard Cite
In this paper, the exponential stability of a kind of It Markovian jump system is studied. Specifically, the exponential stability is ensured by designing an asynchronous delayed‐feedback controller, whose modes are inconsistent with those of the original system. We have built the necessary new relationship between controller modes and system modes, which can facilitate the stability analysis and also make the famous existing ones special cases of the proposed one. Moreover, practically ubiquitous controller delays, which occur between the sensors and controllers, to the detriment of the system stability, are taken into account, and their maximum values are provided in a theoretic way. In this way, the exponential stability can be well ensured against controller delays. With an eye to practical applications, the transition rates of the considered system and the designed asynchronous controller are assumed to be uncertain, and they can be well dealt with by using a new approach with lower conservatism. Furthermore, a practical example and some comparative examples are given to demonstrate the effectiveness and superiorities of the obtained conclusion.
show abstract

Section: Resultsmentioning
confidence: 99%

Section: Introductionmentioning
confidence: 99%

See 2 more Smart Citations
Hidden Markov model‐based asynchronous exponential stabilization for stochastic Markovian jump systems with incomplete controller mode information
Lu,
Li,
Tian
2023
Asian Journal of Control
1
0
0
0
View full text Add to dashboard Cite
In this paper, the exponential stability of a kind of It Markovian jump system is studied. Specifically, the exponential stability is ensured by designing an asynchronous delayed‐feedback controller, whose modes are inconsistent with those of the original system. We have built the necessary new relationship between controller modes and system modes, which can facilitate the stability analysis and also make the famous existing ones special cases of the proposed one. Moreover, practically ubiquitous controller delays, which occur between the sensors and controllers, to the detriment of the system stability, are taken into account, and their maximum values are provided in a theoretic way. In this way, the exponential stability can be well ensured against controller delays. With an eye to practical applications, the transition rates of the considered system and the designed asynchronous controller are assumed to be uncertain, and they can be well dealt with by using a new approach with lower conservatism. Furthermore, a practical example and some comparative examples are given to demonstrate the effectiveness and superiorities of the obtained conclusion.
show abstract

Adaptive Evolutionary Reinforcement Learning with Policy Direction
Dong,
Li
2024
Neural Process Lett
0
0
0
0
View full text Add to dashboard Cite

Evolutionary Reinforcement Learning (ERL) has garnered widespread attention in recent years due to its inherent robustness and parallelism. However, the integration of Evolutionary Algorithms (EAs) and Reinforcement Learning (RL) remains relatively rudimentary and lacks dynamism, which can impact the convergence performance of ERL algorithms. In this study, a dynamic adaptive module is introduced to balance the Evolution Strategies (ES) and RL training within ERL. By incorporating elite strategies, this module leverages advantageous individuals to elevate the overall population's performance. Additionally, RL strategy updates often lack guidance from the population. To address this, we incorporate the strategies of the best individuals from the population, providing valuable policy direction. This is achieved through the formulation of a loss function that employs either L1 or L2 regularization to facilitate RL training. The proposed framework is referred to as Adaptive Evolutionary Reinforcement Learning (AERL). The effectiveness of our framework is evaluated by adopting Soft Actor-Critic (SAC) as the RL algorithm and comparing it with other algorithms in the MuJoCo environment. The results underscore the outstanding convergence performance of our proposed Adaptive Evolutionary Soft Actor-Critic (AESAC) algorithm. Furthermore, ablation experiments are conducted to emphasize the necessity of these two improvements. It is worth noting that the enhancements in AESAC are realized at the population level, enabling broader exploration and effectively reducing the risk of falling into local optima.

show abstract

Zero-sum game-based optimal control for discrete-time Markov jump systems: A parallel off-policy Q-learning method
Wang,
Fang,
Kong
et al. 2024
Applied Mathematics and Computation
0
0
0
0
View full text Add to dashboard Cite

Base on -Learning Pareto Optimality for Linear Itô Stochastic Systems With Markovian Jumps

Cited by 9 publications

References 33 publications

Hidden Markov model‐based asynchronous exponential stabilization for stochastic Markovian jump systems with incomplete controller mode information

Hidden Markov model‐based asynchronous exponential stabilization for stochastic Markovian jump systems with incomplete controller mode information

Adaptive Evolutionary Reinforcement Learning with Policy Direction

Zero-sum game-based optimal control for discrete-time Markov jump systems: A parallel off-policy Q-learning method

Contact Info

Product

Resources

About