ترغب بنشر مسار تعليمي؟ اضغط هنا

Encoding Distributional Soft Actor-Critic for Autonomous Driving in Multi-lane Scenarios

282   0   0.0 ( 0 )
 نشر من قبل Jingliang Duan
 تاريخ النشر 2021
  مجال البحث الهندسة المعلوماتية
والبحث باللغة English




اسأل ChatGPT حول البحث

In this paper, we propose a new reinforcement learning (RL) algorithm, called encoding distributional soft actor-critic (E-DSAC), for decision-making in autonomous driving. Unlike existing RL-based decision-making methods, E-DSAC is suitable for situations where the number of surrounding vehicles is variable and eliminates the requirement for manually pre-designed sorting rules, resulting in higher policy performance and generality. We first develop an encoding distributional policy iteration (DPI) framework by embedding a permutation invariant module, which employs a feature neural network (NN) to encode the indicators of each vehicle, in the distributional RL framework. The proposed DPI framework is proved to exhibit important properties in terms of convergence and global optimality. Next, based on the developed encoding DPI framework, we propose the E-DSAC algorithm by adding the gradient-based update rule of the feature NN to the policy evaluation process of the DSAC algorithm. Then, the multi-lane driving task and the corresponding reward function are designed to verify the effectiveness of the proposed algorithm. Results show that the policy learned by E-DSAC can realize efficient, smooth, and relatively safe autonomous driving in the designed scenario. And the final policy performance learned by E-DSAC is about three times that of DSAC. Furthermore, its effectiveness has also been verified in real vehicle experiments.



قيم البحث

اقرأ أيضاً

Merging into the highway from the on-ramp is an essential scenario for automated driving. The decision-making under the scenario needs to balance the safety and efficiency performance to optimize a long-term objective, which is challenging due to the dynamic, stochastic, and adversarial characteristics. The Rule-based methods often lead to conservative driving on this task while the learning-based methods have difficulties meeting the safety requirements. In this paper, we propose an RL-based end-to-end decision-making method under a framework of offline training and online correction, called the Shielded Distributional Soft Actor-critic (SDSAC). The SDSAC adopts the policy evaluation with safety consideration and a safety shield parameterized with the barrier function in its offline training and online correction, respectively. These two measures support each other for better safety while not damaging the efficiency performance severely. We verify the SDSAC on an on-ramp merge scenario in simulation. The results show that the SDSAC has the best safety performance compared to baseline algorithms and achieves efficient driving simultaneously.
Both single-agent and multi-agent actor-critic algorithms are an important class of Reinforcement Learning algorithms. In this work, we propose three fully decentralized multi-agent natural actor-critic (MAN) algorithms. The agents objective is to co llectively learn a joint policy that maximizes the sum of averaged long-term returns of these agents. In the absence of a central controller, agents communicate the information to their neighbors via a time-varying communication network while preserving privacy. We prove the convergence of all the 3 MAN algorithms to a globally asymptotically stable point of the ODE corresponding to the actor update; these use linear function approximations. We use the Fisher information matrix to obtain the natural gradients. The Fisher information matrix captures the curvature of the Kullback-Leibler (KL) divergence between polices at successive iterates. We also show that the gradient of this KL divergence between policies of successive iterates is proportional to the objective functions gradient. Our MAN algorithms indeed use this emph{representation} of the objective functions gradient. Under certain conditions on the Fisher information matrix, we prove that at each iterate, the optimal value via MAN algorithms can be better than that of the multi-agent actor-critic (MAAC) algorithm using the standard gradients. To validate the usefulness of our proposed algorithms, we implement all the 3 MAN algorithms on a bi-lane traffic network to reduce the average network congestion. We observe an almost 25% reduction in the average congestion in 2 MAN algorithms; the average congestion in another MAN algorithm is on par with the MAAC algorithm. We also consider a generic 15 agent MARL; the performance of the MAN algorithms is again as good as the MAAC algorithm. We attribute the better performance of the MAN algorithms to their use of the above representation.
We develop optimal control strategies for Autonomous Vehicles (AVs) that are required to meet complex specifications imposed by traffic laws and cultural expectations of reasonable driving behavior. We formulate these specifications as rules, and spe cify their priorities by constructing a priority structure. We propose a recursive framework, in which the satisfaction of the rules in the priority structure are iteratively relaxed based on their priorities. Central to this framework is an optimal control problem, where convergence to desired states is achieved using Control Lyapunov Functions (CLFs), and safety is enforced through Control Barrier Functions (CBFs). We also show how the proposed framework can be used for after-the-fact, pass / fail evaluation of trajectories - a given trajectory is rejected if we can find a controller producing a trajectory that leads to less violation of the rule priority structure. We present case studies with multiple driving scenarios to demonstrate the effectiveness of the proposed framework.
With the adoption of autonomous vehicles on our roads, we will witness a mixed-autonomy environment where autonomous and human-driven vehicles must learn to co-exist by sharing the same road infrastructure. To attain socially-desirable behaviors, aut onomous vehicles must be instructed to consider the utility of other vehicles around them in their decision-making process. Particularly, we study the maneuver planning problem for autonomous vehicles and investigate how a decentralized reward structure can induce altruism in their behavior and incentivize them to account for the interest of other autonomous and human-driven vehicles. This is a challenging problem due to the ambiguity of a human drivers willingness to cooperate with an autonomous vehicle. Thus, in contrast with the existing works which rely on behavior models of human drivers, we take an end-to-end approach and let the autonomous agents to implicitly learn the decision-making process of human drivers only from experience. We introduce a multi-agent variant of the synchronous Advantage Actor-Critic (A2C) algorithm and train agents that coordinate with each other and can affect the behavior of human drivers to improve traffic flow and safety.
We develop optimal control strategies for autonomous vehicles (AVs) that are required to meet complex specifications imposed as rules of the road (ROTR) and locally specific cultural expectations of reasonable driving behavior. We formulate these spe cifications as rules, and specify their priorities by constructing a priority structure, called underline{T}otal underline{OR}der over eunderline{Q}uivalence classes (TORQ). We propose a recursive framework, in which the satisfaction of the rules in the priority structure are iteratively relaxed in reverse order of priority. Central to this framework is an optimal control problem, where convergence to desired states is achieved using Control Lyapunov Functions (CLFs) and clearance with other road users is enforced through Control Barrier Functions (CBFs). We present offline and online approaches to this problem. In the latter, the AV has limited sensing range that affects the activation of the rules, and the control is generated using a receding horizon (Model Predictive Control, MPC) approach. We also show how the offline method can be used for after-the-fact (offline) pass/fail evaluation of trajectories - a given trajectory is rejected if we can find a controller producing a trajectory that leads to less violation of the rule priority structure. We present case studies with multiple driving scenarios to demonstrate the effectiveness of the algorithms, and to compare the offline and onli
التعليقات
جاري جلب التعليقات جاري جلب التعليقات
سجل دخول لتتمكن من متابعة معايير البحث التي قمت باختيارها
mircosoft-partner

هل ترغب بارسال اشعارات عن اخر التحديثات في شمرا-اكاديميا