In the application scenarios of quadrotors, it is expected that only part of the obstacles can be identified and located in advance. In order to make quadrotors fly safely in this situation, we present a deep reinforcement learning-based framework to realize autonomous navigation in semi-known environments. Specifically, the proposed framework utilizes the dueling double deep recurrent Q-learning, which can implement global path planning with the obstacle map as input. Moreover, the proposed framework combined with contrastive learning-based feature extraction can conduct real-time autonomous obstacle avoidance with monocular vision effectively. The experimental results demonstrate that our framework exhibits remarkable performance for both global path planning and autonomous obstacle avoidance.

This paper proposes a reinforcement learning (RL) based path following strategy for underactuated airships with magnitude and rate saturation. The Markov decision process (MDP) model for the control problem is established. Then an error bounded line-of-sight (LOS) guidance law is investigated to restrain the state space. Subsequently, a proximal policy optimization (PPO) algorithm is employed to approximate the optimal action policy through trial and error. Since the optimal action policy is generated from the action space, the magnitude and rate saturation can be avoided. The simulation results, involving circular, general, broken-line, and anti-wind path following tasks, demonstrate that the proposed control scheme can transfer to new tasks without adaptation, and possesses satisfying real-time performance and robustness.