As safety violations can lead to severe consequences in real-world robotic applications, the increasing deployment of Reinforcement Learning (RL) in robotic domains has propelled the study of safe exploration for reinforcement learning (safe RL). In this work, we propose a risk preventive training method for safe RL, which learns a statistical contrastive classifier to predict the probability of a state-action pair leading to unsafe states. Based on the predicted risk probabilities, we can collect risk preventive trajectories and reshape the reward function with risk penalties to induce safe RL policies. We conduct experiments in robotic simulation environments. The results show the proposed approach has comparable performance with the state-of-the-art model-based methods and outperforms conventional model-free safe RL approaches.
This paper proposes an on-policy reinforcement learning (RL) control algorithm that solves the optimal regulation problem for a class of uncertain continuous-time nonlinear systems under user-defined state constraints. We formulate the safe RL problem as the minimization of the Hamiltonian subject to a constraint on the time-derivative of a barrier Lyapunov function (BLF). We subsequently use the analytical solution of the optimization problem to modify the Actor-Critic-Identifier architecture to learn the optimal control policy safely. The proposed method does not require the presence of external backup controllers, and the RL policy ensures safety for the entire duration. The efficacy of the proposed controller is demonstrated on a class of Euler-Lagrange systems.
Safe reinforcement learning (RL) studies problems where an intelligent agent has to not only maximize reward but also avoid exploring unsafe areas. In this study, we propose CUP, a novel policy optimization method based on Constrained Update Projection framework that enjoys rigorous safety guarantee. Central to our CUP development is the newly proposed surrogate functions along with the performance bound. Compared to previous safe RL methods, CUP enjoys the benefits of 1) CUP generalizes the surrogate functions to generalized advantage estimator (GAE), leading to strong empirical performance. 2) CUP unifies performance bounds, providing a better understanding and interpretability for some existing algorithms; 3) CUP provides a non-convex implementation via only first-order optimizers, which does not require any strong approximation on the convexity of the objectives. To validate our CUP method, we compared CUP against a comprehensive list of safe RL baselines on a wide range of tasks. Experiments show the effectiveness of CUP both in terms of reward and safety constraint satisfaction. We have opened the source code of CUP at this link https://github.com/zmsn-2077/ CUP-safe-rl.
Prior work on safe Reinforcement Learning (RL) has studied risk-aversion to randomness in dynamics (aleatory) and to model uncertainty (epistemic) in isolation. We propose and analyze a new framework to jointly model the risk associated with epistemic and aleatory uncertainties in finite-horizon and discounted infinite-horizon MDPs. We call this framework that combines Risk-Averse and Soft-Robust methods RASR. We show that when the risk-aversion is defined using either EVaR or the entropic risk, the optimal policy in RASR can be computed efficiently using a new dynamic program formulation with a time-dependent risk level. As a result, the optimal risk-averse policies are deterministic but time-dependent, even in the infinite-horizon discounted setting. We also show that particular RASR objectives reduce to risk-averse RL with mean posterior transition probabilities. Our empirical results show that our new algorithms consistently mitigate uncertainty as measured by EVaR and other standard risk measures.
Reinforcement learning shows great potential to solve complex contact-rich robot manipulation tasks. However, the safety of using RL in the real world is a crucial problem, since unexpected dangerous collisions might happen when the RL policy is imperfect during training or in unseen scenarios. In this paper, we propose a contact-safe reinforcement learning framework for contact-rich robot manipulation, which maintains safety in both the task space and joint space. When the RL policy causes unexpected collisions between the robot arm and the environment, our framework is able to immediately detect the collision and ensure the contact force to be small. Furthermore, the end-effector is enforced to perform contact-rich tasks compliantly, while keeping robust to external disturbances. We train the RL policy in simulation and transfer it to the real robot. Real world experiments on robot wiping tasks show that our method is able to keep the contact force small both in task space and joint space even when the policy is under unseen scenario with unexpected collision, while rejecting the disturbances on the main task.
Optimizing noisy functions online, when evaluating the objective requires experiments on a deployed system, is a crucial task arising in manufacturing, robotics and many others. Often, constraints on safe inputs are unknown ahead of time, and we only obtain noisy information, indicating how close we are to violating the constraints. Yet, safety must be guaranteed at all times, not only for the final output of the algorithm. We introduce a general approach for seeking a stationary point in high dimensional non-linear stochastic optimization problems in which maintaining safety during learning is crucial. Our approach called LB-SGD is based on applying stochastic gradient descent (SGD) with a carefully chosen adaptive step size to a logarithmic barrier approximation of the original problem. We provide a complete convergence analysis of non-convex, convex, and strongly-convex smooth constrained problems, with first-order and zeroth-order feedback. Our approach yields efficient updates and scales better with dimensionality compared to existing approaches. We empirically compare the sample complexity and the computational cost of our method with existing safe learning approaches. Beyond synthetic benchmarks, we demonstrate the effectiveness of our approach on minimizing constraint violation in policy search tasks in safe reinforcement learning (RL).
The action governor is an add-on scheme to a nominal control loop that monitors and adjusts the control actions to enforce safety specifications expressed as pointwise-in-time state and control constraints. In this paper, we introduce the Robust Action Governor (RAG) for systems the dynamics of which can be represented using discrete-time Piecewise Affine (PWA) models with both parametric and additive uncertainties and subject to non-convex constraints. We develop the theoretical properties and computational approaches for the RAG. After that, we introduce the use of the RAG for realizing safe Reinforcement Learning (RL), i.e., ensuring all-time constraint satisfaction during online RL exploration-and-exploitation process. This development enables safe real-time evolution of the control policy and adaptation to changes in the operating environment and system parameters (due to aging, damage, etc.). We illustrate the effectiveness of the RAG in constraint enforcement and safe RL using the RAG by considering their applications to a soft-landing problem of a mass-spring-damper system.
As an important framework for safe Reinforcement Learning, the Constrained Markov Decision Process (CMDP) has been extensively studied in the recent literature. However, despite the rich results under various on-policy learning settings, there still lacks some essential understanding of the offline CMDP problems, in terms of both the algorithm design and the information theoretic sample complexity lower bound. In this paper, we focus on solving the CMDP problems where only offline data are available. By adopting the concept of the single-policy concentrability coefficient $C^*$, we establish an $\Omega\left(\frac{\min\left\{|\mathcal{S}||\mathcal{A}|,|\mathcal{S}|+I\right\} C^*}{(1-\gamma)^3\epsilon^2}\right)$ sample complexity lower bound for the offline CMDP problem, where $I$ stands for the number of constraints. By introducing a simple but novel deviation control mechanism, we propose a near-optimal primal-dual learning algorithm called DPDL. This algorithm provably guarantees zero constraint violation and its sample complexity matches the above lower bound except for an $\tilde{\mathcal{O}}((1-\gamma)^{-1})$ factor. Comprehensive discussion on how to deal with the unknown constant $C^*$ and the potential asynchronous structure on the offline dataset are also included.
Reinforcement Learning (RL) has become an increasingly important research area as the success of machine learning algorithms and methods grows. To combat the safety concerns surrounding the freedom given to RL agents while training, there has been an increase in work concerning Safe Reinforcement Learning (SRL). However, these new and safe methods have been held to less scrutiny than their unsafe counterparts. For instance, comparisons among safe methods often lack fair evaluation across similar initial condition bounds and hyperparameter settings, use poor evaluation metrics, and cherry-pick the best training runs rather than averaging over multiple random seeds. In this work, we conduct an ablation study using evaluation best practices to investigate the impact of run time assurance (RTA), which monitors the system state and intervenes to assure safety, on effective learning. By studying multiple RTA approaches in both on-policy and off-policy RL algorithms, we seek to understand which RTA methods are most effective, whether the agents become dependent on the RTA, and the importance of reward shaping versus safe exploration in RL agent training. Our conclusions shed light on the most promising directions of SRL, and our evaluation methodology lays the groundwork for creating better comparisons in future SRL work.
Reinforcement learning (RL) is a promising optimal control technique for multi-energy management systems. It does not require a model a priori - reducing the upfront and ongoing project-specific engineering effort and is capable of learning better representations of the underlying system dynamics. However, vanilla RL does not provide constraint satisfaction guarantees - resulting in various potentially unsafe interactions within its safety-critical environment. In this paper, we present two novel safe RL methods, namely SafeFallback and GiveSafe, where the safety constraint formulation is decoupled from the RL formulation. These provide hard-constraint, rather than soft- and chance-constraint, satisfaction guarantees both during training a (near) optimal policy (which involves exploratory and exploitative, i.e. greedy, steps) as well as during deployment of any policy (e.g. random agents or offline trained RL agents). This without the need of solving a mathematical program, resulting in less computational power requirements and a more flexible constraint function formulation (no derivative information is required). In a simulated multi-energy systems case study we have shown that both methods start with a significantly higher utility (i.e. useful policy) compared to a vanilla RL benchmark and Optlayer benchmark (94,6% and 82,8% compared to 35,5% and 77,8%) and that the proposed SafeFallback method even can outperform the vanilla RL benchmark (102,9% to 100%). We conclude that both methods are viably safety constraint handling techniques applicable beyond RL, as demonstrated with random policies while still providing hard-constraint guarantees.
Reinforcement Learning (RL) algorithms show amazing performance in recent years, but placing RL in real-world applications such as self-driven vehicles may suffer safety problems. A self-driven vehicle moving to a target position following a learned policy may suffer a vehicle with unpredictable aggressive behaviors or even being pursued by a vehicle following a Nash strategy. To address the safety issue of the self-driven vehicle in this scenario, this paper conducts a preliminary study based on a system of robots. A safe RL framework with safety guarantees is developed for a robot being pursued but with objectives covering more than capture-avoidance. Simulations and experiments are conducted based on the system of robots to evaluate the effectiveness of the developed safe RL framework.
Ensuring safety is a crucial challenge when deploying reinforcement learning (RL) to real-world systems. We develop confidence-based safety filters, a control-theoretic approach for certifying state safety constraints for nominal policies learned via standard RL techniques, based on probabilistic dynamics models. Our approach is based on a reformulation of state constraints in terms of cost functions, reducing safety verification to a standard RL task. By exploiting the concept of hallucinating inputs, we extend this formulation to determine a "backup" policy that is safe for the unknown system with high probability. Finally, the nominal policy is minimally adjusted at every time step during a roll-out towards the backup policy, such that safe recovery can be guaranteed afterwards. We provide formal safety guarantees, and empirically demonstrate the effectiveness of our approach.
Safe reinforcement learning (RL) has achieved significant success on risk-sensitive tasks and shown promise in autonomous driving (AD) as well. Considering the distinctiveness of this community, efficient and reproducible baselines are still lacking for safe AD. In this paper, we release SafeRL-Kit to benchmark safe RL methods for AD-oriented tasks. Concretely, SafeRL-Kit contains several latest algorithms specific to zero-constraint-violation tasks, including Safety Layer, Recovery RL, off-policy Lagrangian method, and Feasible Actor-Critic. In addition to existing approaches, we propose a novel first-order method named Exact Penalty Optimization (EPO) and sufficiently demonstrate its capability in safe AD. All algorithms in SafeRL-Kit are implemented (i) under the off-policy setting, which improves sample efficiency and can better leverage past logs; (ii) with a unified learning framework, providing off-the-shelf interfaces for researchers to incorporate their domain-specific knowledge into fundamental safe RL methods. Conclusively, we conduct a comparative evaluation of the above algorithms in SafeRL-Kit and shed light on their efficacy for safe autonomous driving. The source code is available at \href{ https://github.com/zlr20/saferl_kit}{this https URL}.
Though deep reinforcement learning (DRL) has obtained substantial success, it may encounter catastrophic failures due to the intrinsic uncertainty of both transition and observation. Most of the existing methods for safe reinforcement learning can only handle transition disturbance or observation disturbance since these two kinds of disturbance affect different parts of the agent; besides, the popular worst-case return may lead to overly pessimistic policies. To address these issues, we first theoretically prove that the performance degradation under transition disturbance and observation disturbance depends on a novel metric of Value Function Range (VFR), which corresponds to the gap in the value function between the best state and the worst state. Based on the analysis, we adopt conditional value-at-risk (CVaR) as an assessment of risk and propose a novel reinforcement learning algorithm of CVaR-Proximal-Policy-Optimization (CPPO) which formalizes the risk-sensitive constrained optimization problem by keeping its CVaR under a given threshold. Experimental results show that CPPO achieves a higher cumulative reward and is more robust against both observation and transition disturbances on a series of continuous control tasks in MuJoCo.
Safe reinforcement learning (RL) trains a policy to maximize the task reward while satisfying safety constraints. While prior works focus on the performance optimality, we find that the optimal solutions of many safe RL problems are not robust and safe against carefully designed observational perturbations. We formally analyze the unique properties of designing effective observational adversarial attackers in the safe RL setting. We show that baseline adversarial attack techniques for standard RL tasks are not always effective for safe RL and propose two new approaches - one maximizes the cost and the other maximizes the reward. One interesting and counter-intuitive finding is that the maximum reward attack is strong, as it can both induce unsafe behaviors and make the attack stealthy by maintaining the reward. We further propose a robust training framework for safe RL and evaluate it via comprehensive experiments. This paper provides a pioneer work to investigate the safety and robustness of RL under observational attacks for future safe RL studies. Code is available at: \url{https://github.com/liuzuxin/safe-rl-robustness}
Safe reinforcement learning aims to learn the optimal policy while satisfying safety constraints, which is essential in real-world applications. However, current algorithms still struggle for efficient policy updates with hard constraint satisfaction. In this paper, we propose Penalized Proximal Policy Optimization (P3O), which solves the cumbersome constrained policy iteration via a single minimization of an equivalent unconstrained problem. Specifically, P3O utilizes a simple-yet-effective penalty function to eliminate cost constraints and removes the trust-region constraint by the clipped surrogate objective. We theoretically prove the exactness of the proposed method with a finite penalty factor and provide a worst-case analysis for approximate error when evaluated on sample trajectories. Moreover, we extend P3O to more challenging multi-constraint and multi-agent scenarios which are less studied in previous work. Extensive experiments show that P3O outperforms state-of-the-art algorithms with respect to both reward improvement and constraint satisfaction on a set of constrained locomotive tasks.
Reinforcement Learning (RL) has achieved tremendous success in many complex decision-making tasks. However, safety concerns are raised during deploying RL in real-world applications, leading to a growing demand for safe RL algorithms, such as in autonomous driving and robotics scenarios. While safe control has a long history, the study of safe RL algorithms is still in the early stages. To establish a good foundation for future safe RL research, in this paper, we provide a review of safe RL from the perspectives of methods, theories, and applications. Firstly, we review the progress of safe RL from five dimensions and come up with five crucial problems for safe RL being deployed in real-world applications, coined as "2H3W". Secondly, we analyze the algorithm and theory progress from the perspectives of answering the "2H3W" problems. Particularly, the sample complexity of safe RL algorithms is reviewed and discussed, followed by an introduction to the applications and benchmarks of safe RL algorithms. Finally, we open the discussion of the challenging problems in safe RL, hoping to inspire future research on this thread. To advance the study of safe RL algorithms, we release an open-sourced repository containing the implementations of major safe RL algorithms at the link: https://github.com/chauncygu/Safe-Reinforcement-Learning-Baselines.git.
Constrained reinforcement learning (CRL) has gained significant interest recently, since safety constraints satisfaction is critical for real-world problems. However, existing CRL methods constraining discounted cumulative costs generally lack rigorous definition and guarantee of safety. In contrast, in the safe control research, safety is defined as persistently satisfying certain state constraints. Such persistent safety is possible only on a subset of the state space, called feasible set, where an optimal largest feasible set exists for a given environment. Recent studies incorporate feasible sets into CRL with energy-based methods such as control barrier function (CBF), safety index (SI), and leverage prior conservative estimations of feasible sets, which harms the performance of the learned policy. To deal with this problem, this paper proposes the reachability CRL (RCRL) method by using reachability analysis to establish the novel self-consistency condition and characterize the feasible sets. The feasible sets are represented by the safety value function, which is used as the constraint in CRL. We use the multi-time scale stochastic approximation theory to prove that the proposed algorithm converges to a local optimum, where the largest feasible set can be guaranteed. Empirical results on different benchmarks validate the learned feasible set, the policy performance, and constraint satisfaction of RCRL, compared to CRL and safe control baselines.
Ensuring the safety of reinforcement learning (RL) algorithms is crucial to unlock their potential for many real-world tasks. However, vanilla RL and most safe RL approaches do not guarantee safety. In recent years, several methods have been proposed to provide hard safety guarantees for RL, which is essential for applications where unsafe actions could have disastrous consequences. Nevertheless, there is no comprehensive comparison of these provably safe RL methods. Therefore, we introduce a categorization of existing provably safe RL methods, present the conceptual foundations for both continuous and discrete action spaces, and empirically benchmark existing methods. We categorize the methods based on how they adapt the action: action replacement, action projection, and action masking. Our experiments on an inverted pendulum and a quadrotor stabilization task indicate that action replacement is the best-performing approach for these applications despite its comparatively simple realization. Furthermore, adding a reward penalty, every time the safety verification is engaged, improved training performance in our experiments. Finally, we provide practical guidance on selecting provably safe RL approaches depending on the safety specification, RL algorithm, and type of action space.
Future power systems will rely heavily on micro grids with a high share of decentralised renewable energy sources and energy storage systems. The high complexity and uncertainty in this context might make conventional power dispatch strategies infeasible. Reinforcement-learning based (RL) controllers can address this challenge, however, cannot themselves provide safety guarantees, preventing their deployment in practice. To overcome this limitation, we propose a formally validated RL controller for economic dispatch. We extend conventional constraints by a time-dependent constraint encoding the islanding contingency. The contingency constraint is computed using set-based backwards reachability analysis and actions of the RL agent are verified through a safety layer. Unsafe actions are projected into the safe action space while leveraging constrained zonotope set representations for computational efficiency. The developed approach is demonstrated on a residential use case using real-world measurements.