arXiv subjects

Pavel Osinenko

Publications and source records attributed to Pavel Osinenko.

At least 37 records · Page 2Linked to original sources

arXiv2025

Some remarks on practical stabilization via CLF-based control under measurement noise

Practical stabilization of input-affine systems in the presence of measurement errors and input constraints is considered in this brief note. Assuming that a Lyapunov function and a stabilizing control exist for an input-affine system, the required measurement accuracy at each point of the state space is computed. This is done via the Lyapunov function-based decay condition, which describes along with the input constraints a set of admissible controls. Afterwards, the measurement time points are computed based on the system dynamics. It is shown that between these self-triggered measurement time points, the system evolves and converges into the so-called target ball, i.e. a vicinity of the origin, where it remains. Furthermore, it is shown that the approach ensures the existence of a control law, which is admissible for all possible states and it introduces a connection between measurement time points, measurement accuracy, target ball, and decay. The results of the approach are shown in three examples.

eess.SY↗

arXiv2024

Comprehensive Overview of Reward Engineering and Shaping in Advancing Reinforcement Learning Applications

The aim of Reinforcement Learning (RL) in real-world applications is to create systems capable of making autonomous decisions by learning from their environment through trial and error. This paper emphasizes the importance of reward engineering and reward shaping in enhancing the efficiency and effectiveness of reinforcement learning algorithms. Reward engineering involves designing reward functions that accurately reflect the desired outcomes, while reward shaping provides additional feedback to guide the learning process, accelerating convergence to optimal policies. Despite significant advancements in reinforcement learning, several limitations persist. One key challenge is the sparse and delayed nature of rewards in many real-world scenarios, which can hinder learning progress. Additionally, the complexity of accurately modeling real-world environments and the computational demands of reinforcement learning algorithms remain substantial obstacles. On the other hand, recent advancements in deep learning and neural networks have significantly improved the capability of reinforcement learning systems to handle high-dimensional state and action spaces, enabling their application to complex tasks such as robotics, autonomous driving, and game playing. This paper provides a comprehensive review of the current state of reinforcement learning, focusing on the methodologies and techniques used in reward engineering and reward shaping. It critically analyzes the limitations and recent advancements in the field, offering insights into future research directions and potential applications in various domains.

cs.LG↗

arXiv2024

Optimizing energy consumption for legged robot by adapting equilibrium position and stiffness of a parallel torsion spring

This paper is dedicated to the development of a novel adaptive torsion spring mechanism for optimizing energy consumption in legged robots. By adjusting the equilibrium position and stiffness of the spring, the system improves energy efficiency during cyclic movements, such as walking and jumping. The adaptive compliance mechanism, consisting of a torsion spring combined with a worm gear driven by a servo actuator, compensates for motion-induced torque and reduces motor load. Simulation results demonstrate a significant reduction in power consumption, highlighting the effectiveness of this approach in enhancing robotic locomotion.

cs.RO↗

arXiv2024

A novel agent with formal goal-reaching guarantees: an experimental study with a mobile robot

Reinforcement Learning (RL) has been shown to be effective and convenient for a number of tasks in robotics. However, it requires the exploration of a sufficiently large number of state-action pairs, many of which may be unsafe or unimportant. For instance, online model-free learning can be hazardous and inefficient in the absence of guarantees that a certain set of desired states will be reached during an episode. An increasingly common approach to address safety involves the addition of a shielding system that constrains the RL actions to a safe set of actions. In turn, a difficulty for such frameworks is how to effectively couple RL with the shielding system to make sure the exploration is not excessively restricted. This work presents a novel safe model-free RL agent called Critic As Lyapunov Function (CALF) and showcases how CALF can be used to improve upon control baselines in robotics in an efficient and convenient fashion while ensuring guarantees of stable goal reaching. The latter is a crucial part of safety, as seen generally. With CALF all state-action pairs remain explorable and yet reaching of desired goal states is formally guaranteed. Formal analysis is provided that shows the goal stabilization-ensuring properties of CALF and a set of real-world and numerical experiments with a non-holonomic wheeled mobile robot (WMR) TurtleBot3 Burger confirmed the superiority of CALF over such a well-established RL agent as proximal policy optimization (PPO), and a modified version of SARSA in a few-episode setting in terms of attained total cost.

cs.RO↗

arXiv2024

Critic as Lyapunov function (CALF): a model-free, stability-ensuring agent

This work presents and showcases a novel reinforcement learning agent called Critic As Lyapunov Function (CALF) which is model-free and ensures online environment, in other words, dynamical system stabilization. Online means that in each learning episode, the said environment is stabilized. This, as demonstrated in a case study with a mobile robot simulator, greatly improves the overall learning performance. The base actor-critic scheme of CALF is analogous to SARSA. The latter did not show any success in reaching the target in our studies. However, a modified version thereof, called SARSA-m here, did succeed in some learning scenarios. Still, CALF greatly outperformed the said approach. CALF was also demonstrated to improve a nominal stabilizer provided to it. In summary, the presented agent may be considered a viable approach to fusing classical control with reinforcement learning. Its concurrent approaches are mostly either offline or model-based, like, for instance, those that fuse model-predictive control into the agent.

cs.RO↗

arXiv2024

An agent design with goal reaching guarantees for enhancement of learning

Reinforcement learning is commonly concerned with problems of maximizing accumulated rewards in Markov decision processes. Oftentimes, a certain goal state or a subset of the state space attain maximal reward. In such a case, the environment may be considered solved when the goal is reached. Whereas numerous techniques, learning or non-learning based, exist for solving environments, doing so optimally is the biggest challenge. Say, one may choose a reward rate which penalizes the action effort. Reinforcement learning is currently among the most actively developed frameworks for solving environments optimally by virtue of maximizing accumulated reward, in other words, returns. Yet, tuning agents is a notoriously hard task as reported in a series of works. Our aim here is to help the agent learn a near-optimal policy efficiently while ensuring a goal reaching property of some basis policy that merely solves the environment. We suggest an algorithm, which is fairly flexible, and can be used to augment practically any agent as long as it comprises of a critic. A formal proof of a goal reaching property is provided. Comparative experiments on several problems under popular baseline agents provided an empirical evidence that the learning can indeed be boosted while ensuring goal reaching property.

cs.AI↗

arXiv2024

On constructive extractability of measurable selectors of set-valued maps

This paper investigates the possibility of constructive extraction of measurable selector from set-valued maps which may commonly arise in viability theory, optimal control, discontinuous systems etc. For instance, existence of solutions to certain differential inclusions, often requires iterative extraction of measurable selectors. Next, optimal controls are in general non-unique which naturally leads to an optimal set-valued function. Finally, a viable control law can be seen, in general, as a selector. It is known that selector theorems are non-constructive and so selectors cannot always be extracted. In this work, we analyze under which particular conditions selectors are constructively extractable. An algorithm is derived from the theorem and applied in a computational study with a three-wheel robot model.

eess.SY↗

arXiv2023

Combining model-predictive control and predictive reinforcement learning for stable quadrupedal robot locomotion

Stable gait generation is a crucial problem for legged robot locomotion as this impacts other critical performance factors such as, e.g. mobility over an uneven terrain and power consumption. Gait generation stability results from the efficient control of the interaction between the legged robot's body and the environment where it moves. Here, we study how this can be achieved by a combination of model-predictive and predictive reinforcement learning controllers. Model-predictive control (MPC) is a well-established method that does not utilize any online learning (except for some adaptive variations) as it provides a convenient interface for state constraints management. Reinforcement learning (RL), in contrast, relies on adaptation based on pure experience. In its bare-bone variants, RL is not always suitable for robots due to their high complexity and expensive simulation/experimentation. In this work, we combine both control methods to address the quadrupedal robot stable gate generation problem. The hybrid approach that we develop and apply uses a cost roll-out algorithm with a tail cost in the form of a Q-function modeled by a neural network; this allows to alleviate the computational complexity, which grows exponentially with the prediction horizon in a purely MPC approach. We demonstrate that our RL gait controller achieves stable locomotion at short horizons, where a nominal MP controller fails. Further, our controller is capable of live operation, meaning that it does not require previous training. Our results suggest that the hybridization of MPC with RL, as presented here, is beneficial to achieve a good balance between online control capabilities and computational complexity.

cs.RO↗

arXiv2023

Experimental verification of an online traction parameter identification method

Traction parameters, that characterize the ground-wheel contact dynamics, are the central factor in the energy efficiency of vehicles. To optimize fuel consumption, reduce wear of tires, increase productivity etc., knowledge of current traction parameters is unavoidable. Unfortunately, these parameters are difficult to measure and require expensive force and torque sensors. An alternative way is to use system identification to determine them. In this work, we validate such a method in field experiments with a mobile robot. The method is based on an adaptive Kalman filter. We show how it estimates the traction parameters online, during the motion on the field, and compare them to their values determined via a 6-directional force-torque sensor installed for verification. Data of adhesion slip ratio curves is recorded and compared to curves from literature for additional validation of the method. The results can establish a foundation for a number of optimal traction methods.

eess.SY↗

arXiv2022

A generalized stacked reinforcement learning method for sampled systems

A common setting of reinforcement learning (RL) is a Markov decision process (MDP) in which the environment is a stochastic discrete-time dynamical system. Whereas MDPs are suitable in such applications as video-games or puzzles, physical systems are time-continuous. A general variant of RL is of digital format, where updates of the value (or cost) and policy are performed at discrete moments in time. The agent-environment loop then amounts to a sampled system, whereby sample-and-hold is a specific case. In this paper, we propose and benchmark two RL methods suitable for sampled systems. Specifically, we hybridize model-predictive control (MPC) with critics learning the optimal Q- and value (or cost-to-go) function. Optimality is analyzed and performance comparison is done in an experimental case study with a mobile robot.

cs.RO↗

arXiv2022

A framework for online, stabilizing reinforcement learning

Online reinforcement learning is concerned with training an agent on-the-fly via dynamic interaction with the environment. Here, due to the specifics of the application, it is not generally possible to perform long pre-training, as it is commonly done in off-line, model-free approaches, which are akin to dynamic programming. Such applications may be found more frequently in industry, rather than in pure digital fields, such as cloud services, video games, database management, etc., where reinforcement learning has been demonstrating success. Online reinforcement learning, in contrast, is more akin to classical control, which utilizes some model knowledge about the environment. Stability of the closed-loop (agent plus the environment) is a major challenge for such online approaches. In this paper, we tackle this problem by a special fusion of online reinforcement learning with elements of classical control, namely, based on the Lyapunov theory of stability. The idea is to start the agent at once, without pre-training, and learn approximately optimal policy under specially designed constraints, which guarantee stability. The resulting approach was tested in an extensive experimental study with a mobile robot. A nominal parking controller was used as a baseline. It was observed that the suggested agent could always successfully park the robot, while significantly improving the cost. While many approaches may be exploited for mobile robot control, we suggest that the experiments showed the promising potential of online reinforcement learning agents based on Lyapunov-like constraints. The presented methodology may be utilized in safety-critical, industrial applications where stability is necessary.

eess.SY↗

arXiv2022

On stochastic stabilization via non-smooth control Lyapunov functions

Control Lyapunov function is a central tool in stabilization. It generalizes an abstract energy function -- a Lyapunov function -- to the case of controlled systems. It is a known fact that most control Lyapunov functions are non-smooth -- so is the case in non-holonomic systems, like wheeled robots and cars. Frameworks for stabilization using non-smooth control Lyapunov functions exist, like Dini aiming and steepest descent. This work generalizes the related results to the stochastic case. As the groundwork, sampled control scheme is chosen in which control actions are computed at discrete moments in time using discrete measurements of the system state. In such a setup, special attention should be paid to the sample-to-sample behavior of the control Lyapunov function. A particular challenge here is a random noise acting on the system. The central result of this work is a theorem that states, roughly, that if there is a, generally non-smooth, control Lyapunov function, the given stochastic dynamical system can be practically stabilized in the sample-and-hold mode meaning that the control actions are held constant within sampling time steps. A particular control method chosen is based on Moreau-Yosida regularization, in other words, inf-convolution of the control Lyapunov function, but the overall framework is extendable to further control schemes. It is assumed that the system noise be bounded almost surely, although the case of unbounded noise is briefly addressed.

math.OC↗

arXiv2022

On stochastic stabilization of sampled systems

This paper addresses stochastic stabilization in case where implementation of control policies is digital, i. e., when the dynamical system is treated continuous, whereas the control actions are held constant in predefined time steps. In such a setup, special attention should be paid to the sample-to-sample behavior of the involved Lyapunov function. This paper extends on the stochastic stability results specifically to address for the sample-and-hold mode. We show that if a Markov policy stabilizes the system in a suitable sense, then it also practically stabilizes it in the sample-and-hold sense. This establishes a bridge from an idealized continuous application of the policy to its digital implementation. The central result applies to dynamical systems described by stochastic differential equations driven by the standard Brownian motion. Generalizations are discussed, including the case of non-smooth Lyapunov functions for systems driven by bounded noise. A brief overview of bounded noise models is given.

math.DS↗

arXiv2022

A stabilizing reinforcement learning approach for sampled systems with partially unknown models

Reinforcement learning is commonly associated with training of reward-maximizing (or cost-minimizing) agents, in other words, controllers. It can be applied in model-free or model-based fashion, using a priori or online collected system data to train involved parametric architectures. In general, online reinforcement learning does not guarantee closed loop stability unless special measures are taken, for instance, through learning constraints or tailored training rules. Particularly promising are hybrids of reinforcement learning with "classical" control approaches. In this work, we suggest a method to guarantee practical stability of the system-controller closed loop in a purely online learning setting, i.e., without offline training. Moreover, we assume only partial knowledge of the system model. To achieve the claimed results, we employ techniques of classical adaptive control. The implementation of the overall control scheme is provided explicitly in a digital, sampled setting. That is, the controller receives the state of the system and computes the control action at discrete, specifically, equidistant moments in time. The method is tested in adaptive traction control and cruise control where it proved to significantly reduce the cost.

eess.SY↗

arXiv2022

A note on stabilizing reinforcement learning

Reinforcement learning is a general methodology of adaptive optimal control that has attracted much attention in various fields ranging from video game industry to robot manipulators. Despite its remarkable performance demonstrations, plain reinforcement learning controllers do not guarantee stability which compromises their applicability in industry. To provide such guarantees, measures have to be taken. This gives rise to what could generally be called stabilizing reinforcement learning. Concrete approaches range from employment of human overseers to filter out unsafe actions to formally verified shields and fusion with classical stabilizing controllers. A line of attack that utilizes elements of adaptive control has become fairly popular in the recent years. In this note, we critically address such an approach in a fairly general actor-critic setup for nonlinear time-continuous environments. The actor network utilizes a so-called robustifying term that is supposed to compensate for the neural network errors. The corresponding stability analysis is based on the value function itself. We indicate a problem in such a stability analysis and provide a counterexample to the overall control scheme. Implications for such a line of attack in stabilizing reinforcement learning are discussed. Furthermore, unfortunately the said problem possess no fix without a substantial reconsideration of the whole approach. As a positive message, we derive a stochastic critic neural network weight convergence analysis provided that the environment was stabilized.

math.DS↗

arXiv2021

Effects of sampling and horizon in predictive reinforcement learning

Plain reinforcement learning (RL) may be prone to loss of convergence, constraint violation, unexpected performance, etc. Commonly, RL agents undergo extensive learning stages to achieve acceptable functionality. This is in contrast to classical control algorithms which are typically model-based. An direction of research is the fusion of RL with such algorithms, especially model-predictive control (MPC). This, however, introduces new hyper-parameters related to the prediction horizon. Furthermore, RL is usually concerned with Markov decision processes. But the most of the real environments are not time-discrete. The factual physical setting of RL consists of a digital agent and a time-continuous dynamical system. There is thus, in fact, yet another hyper-parameter -- the agent sampling time. In this paper, we investigate the effects of prediction horizon and sampling of two hybrid RL-MPC-agents in a case study with a mobile robot parking, which is in turn a canonical control problem. We benchmark the agents with a simple variant of MPC. The sampling showed a kind of a "sweet spot" behavior, whereas the RL agents demonstrated merits at shorter horizons.

math.DS↗

arXiv2021

An experimental study of two predictive reinforcement learning methods and comparison with model-predictive control

Reinforcement learning (RL) has been successfully used in various simulations and computer games. Industry-related applications, such as autonomous mobile robot motion control, are somewhat challenging for RL up to date though. This paper presents an experimental evaluation of predictive RL controllers for optimal mobile robot motion control. As a baseline for comparison, model-predictive control (MPC) is used. Two RL methods are tested: a roll-out Q-learning, which may be considered as MPC with terminal cost being a Q-function approximation, and a so-called stacked Q-learning, which in turn is like MPC with the running cost substituted for a Q-function approximation. The experimental foundation is a mobile robot with a differential drive (Robotis Turtlebot3). Experimental results showed that both RL methods beat the baseline in terms of the accumulated cost, whereas the stacked variant performed best. Provided the series of previous works on stacked Q-learning, this particular study supports the idea that MPC with a running cost adaptation inspired by Q-learning possesses potential of performance boost while retaining the nice properties of MPC.

cs.RO↗

arXiv2021

Optimal control of centrifugal spreader

Achieving an evenly distributed fertilization spread pattern is a complex technical task. A corresponding control algorithm must account for the tractor movement, the settings of the spreader, the prescribed dosage as well as machine constraints. It dictates, in particular, the fertilization process needs be estimated ahead to achieve an optimized spread pattern. The presented work is concerned with the development of a predictive control scheme for optimized fertilizer application using modeling of the tractor moving on the field and the spread pattern in form of a crescent behind the tractor. In particular, the form of the spread pattern is modeled via four normal distributions, two for each side of the pattern. The control goal is to achieve a desired fertilizer distribution on the field. The study presents three algorithms for comparison: a one-step optimization and two approaches using model-predictive control, one with a simplified model of the spread pattern in the prediction horizon, and one with a comprehensive one model, respectively. The best results are obtained with model-predictive control using the comprehensive model.

eess.SY↗