Learning to map the images acquired by a moving agent equipped with a camera sensor to motion commands for multigoal navigation is challenging. Most existing approaches are still struggling against collision avoidance, faster convergence, and generalization. In this article, a novel actor–critic architecture is presented to learn the optimal navigation policy. We introduce single-step reward observation and collision penalty to reshape the reinforcement learning (RL) reward function. The collision perception can be obtained by the reshaped reward function and treated as measurement information from the visual observation to avoid obstacles. Besides, expert trajectories are used to generate subgoals. A subgoal reward shaping is then proposed to accelerate policy learning with the expert knowledge of subgoals. In order to generate human-aware navigation policies, an observation-action consistency (OAC) model is introduced to ensure that the agent reaches the subgoals in turn, and moves toward the target. The whole training process is performed on a self-supervised RL approach, accompanied by an expert supervision signal. This method balances the exploration and exploitation, helping the proposed model to generalize to unseen goals. The training experiments on AI2-THOR show better performance and faster convergence speed, compared with the existing approaches. For the generalization capacity to unseen goals, the proposed method achieves the state-of-the-art success rate, with at least a 30% improvement of average episode collision.

It is still a challenge to realize safe and fast autonomous driving through deep reinforcement learning. Most autonomous driving reinforcement learning models are subject to a single experience replay approach for training agents and how to improve the driving speed and safety of agent has become the focus of research. Therefore, we present an improved Double Bias Experience Replay (DBER) approach, which enables the agent to choose its own driving learning tendency. A new loss function is proposed to ameliorate the relationship between negative loss and positive loss. The proposed approach has been applied to three algorithms to verify: Deep Q Network (DQN), Dueling Double DQN (DD-DQN) and Quantile Regression DQN (QR-DQN). Compared with the existing approaches, the proposed approach show better performance and robustness of driving policy on the driving simulator, which is implemented by the Unity ML-agents. The approach makes the vehicle agent obtain better performance, such as higher reward, faster driving speed, less lane changing and more in the same training time.

The main objective of this paper is to handle the trajectory tracking control for a class of three-wheel mobile robot by using a novel adaptive robust control. The mobile robot consists of nonholonomic constraints and uncertainties. The uncertainties include initial condition offset, mass, and moment of inertia of the system, which are time-varying and bounded (unknown). The control force in analytical form is obtained by UK theory, and the adaptive robust control is formulated to handle the uncertainties. For estimating the unknown bounds of uncertainties, the leakage-type adaptive law is designed, which can automatically adjust its own value according to the trajectory tracking errors. Then we verify the uniform boundedness and uniform ultimate boundedness of the proposed control by Lyapunov method. Finally, numerical simulations are conducted to show the trajectory tracking effectiveness of the designed control method.