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.

Reinforcement learning (RL) has gained wide attention, but its implementation in autonomous vehicles is still limited by insufficient sample efficiency and heavy training costs. The training efficiency of RL agents is influenced by the dimension of the state space, which can be partitioned to reduce the complexity of sampling and computation. This study proposes a hierarchical clustering-based state grouping reinforcement learning (HCSG-RL) method for the switching decision of autonomous vehicles. First, we partition the base state space into groups and generate a hierarchical tree of state space groups. Then, we train multiple sub-agents for each node in the hierarchical tree. Finally, we add these trained-well sub-model into master policy. This method allows us to fully explore all state spaces and improve the training efficiency of individual agents, which handles the 'long-tail' issue and the curse of dimensionality issue. We conduct experiments in a simulation environment and results show that the proposed method has 16-72% reward improvement compared to the tree model in different road length.

Scene text detection has attracted increasing concerns with the rapid development of deep neural networks in recent years. However, existing scene text detectors may overfit on the public datasets due to the limited training data, or generate inaccurate localization for arbitrary-shape scene texts. This paper presents an arbitrary-shape scene text detection method that can achieve better generalization ability and more accurate localization. We first propose a Scale-Aware Data Augmentation (SADA) technique to increase the diversity of training samples. SADA considers the scale variations and local visual variations of scene texts, which can effectively relieve the dilemma of limited training data. At the same time, SADA can enrich the training minibatch, which contributes to accelerating the training process. Furthermore, a Shape Similarity Constraint (SSC) technique is exploited to model the global shape structure of arbitrary-shape scene texts and backgrounds from the perspective of the loss function. SSC encourages the segmentation of text or non-text in the candidate boxes to be similar to the corresponding ground truth, which is helpful to localize more accurate boundaries for arbitrary-shape scene texts. Extensive experiments have demonstrated the effectiveness of the proposed techniques, and state-of-the-art performances are achieved over public arbitrary-shape scene text benchmarks (e.g., CTW1500, Total-Text, and ArT).