Complete stability analysis of a heuristic ADP control design
نویسندگان
چکیده
This paper provides new stability results for Action-Dependent Heuristic Dynamic Programming (ADHDP), using a control algorithm that iteratively improves an internal model of the external world in the autonomous system based on its continuous interaction with the environment. We extend previous results by ADHDP control to the case of general multi-layer neural networks with deep learning across all layers. In particular, we show that the introduced control approach is uniformly ultimately bounded (UUB) under specific conditions on the learning rates, without explicit constraints on the temporal discount factor. We demonstrate the benefit of our results to the control of linear and nonlinear systems, including the cart-pole balancing problem. Our results show significantly improved learning and control performance as compared to the state-of-art.
منابع مشابه
A Preprocessing Technique to Investigate the Stability of Multi-Objective Heuristic Ensemble Classifiers
Background and Objectives: According to the random nature of heuristic algorithms, stability analysis of heuristic ensemble classifiers has particular importance. Methods: The novelty of this paper is using a statistical method consists of Plackett-Burman design, and Taguchi for the first time to specify not only important parameters, but also optimal levels for them. Minitab and Design Expert ...
متن کاملStability and Robust Performance Analysis of Fractional Order Controller over Conventional Controller Design
In this paper, a new comparative approach has been proposed for reliable controller design. Scientists and engineers are often confronted with the analysis, design, and synthesis of real-life problems. The first step in such studies is the development of a 'mathematical model' which can be considered as a substitute for the real problem. The mathematical model is used here as a plant. Fractiona...
متن کاملComplete stability analysis of a heuristic approximate dynamic programming control design
This paper provides new stability results for Action-Dependent Heuristic Dynamic Programming (ADHDP), using a control algorithm that iteratively improves an internal model of the external world in the autonomous system based on its continuous interaction with the environment. We extend previous results for ADHDP control to the case of general multi-layer neural networks with deep learning acros...
متن کاملStability Analysis and Robust Controller Design for Uncertain Discrete-time Singularly Perturbed Systems
In this paper, the stability analysis and controller design for uncertain discretetime singularly perturbed system are investigated via a matrix inequality approach. In analysis, the stability condition under which the singularly perturbed system is quadratically stable for sufficiently small singular perturbation parameter is derived in the formulation of linear matrix inequality (LMI). In syn...
متن کاملتجزیه پایداری ژنوتیپهای جو در آزمایشهای یکنواخت سراسری منطقه سرد
To determine yield stability and to evaluate genotype interaction with environment interaction, 18 genotype of barley (Hordeum vulgare L.) and a control group were evaluated in a randomized complete block design with 4 replications in 3 successive years (1997-2000) at 10 research stations. Simple and combined analysis of variance revealed significant genetic differences between yield genotypes ...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
عنوان ژورنال:
- CoRR
دوره abs/1308.3282 شماره
صفحات -
تاریخ انتشار 2013