This study presents a general optimal trajectory planning (GOTP) framework for autonomous vehicles (AVs) that can effectively avoid obstacles and guide AVs to complete driving tasks safely and efficiently. Firstly, we employ the fifth-order Bezier curve to generate and smooth the reference path along the road centerline. Cartesian coordinates are then transformed to achieve the curvature continuity of the generated curve. Considering the road constraints and vehicle dynamics, limited polynomial candidate trajectories are generated and smoothed in a curvilinear coordinate system. Furthermore, in selecting the optimal trajectory, we develop a unified and auto-tune objective function based on the principle of least action by employing AVs to simulate drivers’ behavior and summarizing their manipulation characteristics of “seeking benefits and avoiding losses.” Finally, by integrating the idea of receding-horizon optimization, the proposed framework is achieved by considering dynamic multi-performance objectives and selecting trajectories that satisfy feasibility, optimality, and adaptability. Extensive simulations and experiments are performed, and the results demonstrate the framework's feasibility and effectiveness, which avoids both dynamic and static obstacles and applies to various scenarios with multi-source interactive traffic participants. Moreover, we prove that the proposed method can guarantee real-time planning and safety requirements compared to drivers’ manipulation.

@en

In traditional reinforcement learning (RL) problems, agents can explore environments to learn optimal policies through trials and errors that are sometimes unsafe. However, unsafe interactions with environments are unacceptable in many safety-critical problems, for instance in robot navigation tasks. Even though RL agents can be trained in simulators, there are many real-world problems without simulators of sufficient fidelity. Constructing safe exploration algorithms for dangerous environments is challenging because we have to optimize policies under the premise of safety. In general, safety is still an open problem that hinders the wider application of RL.@en

Unmanned Aerial Vehicle (UAV) maneuver strategy learning remains a challenge when using Reinforcement Learning (RL) in this sparse reward task. In this paper, we propose Subtask-Masked curriculum learning for RL (SUBMAS-RL), an efficient RL paradigm that implements curriculum learning and knowledge transfer for UAV maneuver scenarios involving multiple missiles. First, this study introduces a novel concept known as subtask mask to create source tasks from a target task by masking partial subtasks. Then, a subtask-masked curriculum generation method is proposed to generate a sequenced curriculum by alternately conducting task generation and task sequencing. To establish efficient knowledge transfer and avoid negative transfer, this paper employs two transfer techniques, policy distillation and policy reuse, along with an explicit transfer condition that masks irrelevant knowledge. Experimental results demonstrate that our method achieves a 94.8% success rate in the UAV maneuver scenario, where the direct use of reinforcement learning always fails. The proposed RL framework SUBMAS-RL is expected to learn an effective policy in complex tasks with sparse rewards.

@en

Safety is critical to broadening the application of reinforcement learning (RL). Often, we train RL agents in a controlled environment, such as a laboratory, before deploying them in the real world. However, the real-world target task might be unknown prior to deployment. Reward-free RL trains an agent without the reward to adapt quickly once the reward is revealed. We consider the constrained reward-free setting, where an agent (the guide) learns to explore safely without the reward signal. This agent is trained in a controlled environment, which allows unsafe interactions and still provides the safety signal. After the target task is revealed, safety violations are not allowed anymore. Thus, the guide is leveraged to compose a safe behaviour policy. Drawing from transfer learning, we also regularize a target policy (the student) towards the guide while the student is unreliable and gradually eliminate the influence of the guide as training progresses. The empirical analysis shows that this method can achieve safe transfer learning and helps the student solve the target task faster.@en

Without an assigned task, a suitable intrinsic objective for an agent is to explore the environment efficiently. However, the pursuit of exploration will inevitably bring more safety risks. An under-explored aspect of reinforcement learning is how to achieve safe efficient exploration when the task is unknown. In this paper, we propose a practical Constrained Entropy Maximization (CEM) algorithm to solve task-agnostic safe exploration problems, which naturally require a finite horizon and undiscounted constraints on safety costs. The CEM algorithm aims to learn a policy that maximizes the state entropy under the premise of safety. To avoid approximating the state density in complex domains, CEM leverages a $k$-nearest neighbor entropy estimator to evaluate the efficiency of exploration. In terms of safety, CEM minimizes the safety costs, and adaptively trades off safety and exploration based on the current constraint satisfaction. We empirically show that CEM allows learning a safe exploration policy in complex continuous-control domains, and the learned policy benefits downstream tasks in safety and sample efficiency.@en

Safety is critical to broadening the a lication of reinforcement learning (RL). Often, RL agents are trained in a controlled environment, such as a laboratory, before being de loyed in the real world. However, the target reward might be unknown rior to de loyment. Reward-free RL addresses this roblem by training an agent without the reward to ada t quickly once the reward is revealed. We consider the constrained reward-free setting, where an agent (the guide) learns to ex lore safely without the reward signal. This agent is trained in a controlled environment, which allows unsafe interactions and still rovides the safety signal. After the target task is revealed, safety violations are not allowed anymore. Thus, the guide is leveraged to com ose a safe sam ling olicy. Drawing from transfer learning, we also regularize a target olicy (the student) towards the guide while the student is unreliable and gradually eliminate the influence from the guide as training rogresses. The em irical analysis shows that this method can achieve safe transfer learning and hel s the student solve the target task faster. @en

The use of reinforcement learning (RL) in real-world domains often requires extensive effort to ensure safe behavior. While this compromises the autonomy of the system, it might still be too risky to allow a learning agent to freely explore its environment. These strict impositions come at the cost of flexibility and applying them often relies on complex parameters and hard-coded knowledge modelled by the reward function. Autonomous driving is one such domain that could greatly benefit from more efficient and verifiable methods for safe automation. We propose to approach the automated driving problem using constrained RL, a method that automates the trade off between risk and utility, thereby significantly reducing the burden on the designer. We first show that an engineered reward function for ensuring safety and utility in one specific environment might not result in the optimal behavior when traffic dynamics changes in the exact environment. Next we show how algorithms based on constrained RL which are more robust to the environmental disturbances can address this challenge. These algorithms use a simple and easy to interpret reward and cost function, and are able to maintain both, efficiency and safety without requiring reward parameter tuning. We demonstrate our approach in the automated merging scenario with different traffic configurations such as low or high chance of cooperative drivers and different cooperative driving strategies.@en

Safety is critical to broadening the real-world use of reinforcement learning (RL). Modeling the safety aspects using a safety-cost signal separate from the reward is becoming standard practice, since it avoids the problem of finding a good balance between safety and performance. However, the total safety-cost distribution of different trajectories is still largely unexplored. In this paper, we propose an actor critic method for safe RL that uses an implicit quantile network to approximate the distribution of accumulated safety-costs. Using an accurate estimate of the distribution of accumulated safetycosts, in particular of the upper tail of the distribution, greatly improves the performance of riskaverse RL agents. The empirical analysis shows that our method achieves good risk control in complex safety-constrained environments. @en

Safety is critical to broadening the real-world use of reinforcement learning. Modeling the safety aspects using a safety-cost signal separate from the reward and bounding the expected safety-cost is becoming standard practice, since it avoids the problem of finding a good balance between safety and performance. However, it can be risky to set constraints only on the expectation neglecting the tail of the distribution, which might have prohibitively large values. In this paper, we propose a method called Worst-Case Soft Actor Critic for safe RL that approximates the distribution of accumulated safety-costs to achieve risk control. More specifically, a certain level of conditional Value-at-Risk from the distribution is regarded as a safety constraint, which guides the change of adaptive safety weights to achieve a trade-off between reward and safety. As a result, we can compute policies whose worst-case performance satisfies the constraints. We investigate two ways to estimate the safety-cost distribution, namely a Gaussian approximation and a quantile regression algorithm. On the one hand, the Gaussian approximation is simple and easy to implement, but may underestimate the safety cost, on the other hand, the quantile regression leads to a more conservative behavior. The empirical analysis shows that the quantile regression method achieves excellent results in complex safety-constrained environments, showing good risk control.@en

Safe exploration is regarded as a key priority area for reinforcement learning research. With separate reward and safety signals, it is natural to cast it as constrained reinforcement learning, where expected long-term costs of policies are constrained. However, it can be hazardous to set constraints on the expected safety signal without considering the tail of the distribution. For instance, in safety-critical domains, worst-case analysis is required to avoid disastrous results. We present a novel reinforcement learning algorithm called Worst-Case Soft Actor Critic, which extends the Soft Actor Critic algorithm with a safety critic to achieve risk control. More specifically, a certain level of conditional Value-at- Risk from the distribution is regarded as a safety measure to judge the constraint satisfaction, which guides the change of adaptive safety weights to achieve a trade-off between reward and safety. As a result, we can optimize policies under the premise that their worst-case performance satisfies the constraints. The empirical analysis shows that our algorithm attains better risk control compared to expectation-based methods. @en