Autonomous navigation of unmanned aerial vehicles (UAVs) in unknown complex environments requires safe, fast and efficient path planning algorithms. Currently, the two-stage framework of “front-end search and back-end optimization” is widely adopted. However, existing research primarily focuses on path planning performance in high-density obstacle scenarios, lacking effective strategies for large-sized obstacles. Furthermore, the current two-stage framework suffers from issues such as path divergence and reduced flight speed. To address these limitations, this paper proposes a general path planning algorithm with soft constraints for UAVs in high-density obstacle scenarios and large-sized obstacle scenarios. The core of the algorithm involves guiding the UAV trajectory through the establishment of well-defined local target points. The front-end employs an expanded space observer for two observations, constructing a real-time safety region, and integrates flight state information to generate local target points using reinforcement learning. The back-end generates trajectories that allows UAVs to fly towards the local target points at higher speeds through an improved Soft Differential Constrained Minimum Snap (SDC-Minimum Snap) algorithm. For large-sized obstacles, a cost-function-based backtracking and circumvention mechanism is introduced to ensure reliable obstacle avoidance. Simulations and real-world experiments validate the generality and feasibility of the proposed algorithm in both scenarios.

Aiming at the problem that autonomous underwater vehicles (AUVs) can only rely on inertial navigation system (INS) for navigation in specific environments (noncooperative environments or extreme cases of external sensor failures, etc.) and that the navigation accuracy will decrease rapidly over time due to the cumulative nature of INS errors in the absence of external signal correction, this article proposes a self-aided navigation strategy. The core idea of this strategy is to deeply explore the characteristics of INS itself by perceiving the AUV’s motion characteristics and environmental changes, extract navigation-related auxiliary information, and fuse this information with the INS to correct the accumulated error of the INS during the AUV’s long navigation process, as well as to provide a reliable priori state for the subsequent navigation. This article combines theoretical analysis, mathematical simulation, and field experiments to comprehensively verify the self-aided navigation strategy under circular motion trajectory constraints. The experimental results show that the strategy realizes the correction of INS attitude and velocity, effectively delays the INS error dispersion, improves the autonomous navigation capability in the absence of external correction signals, and provides new ideas and practical basis for the application of inertial navigation technology in long-endurance navigation specific environments.

To address the challenge of DVL beam failure in complex underwater environments, especially under water tracking mode, a robust SINS/DVL/PS tightly integrated navigation method under partial beam failure is proposed in this paper. A new Doppler shift-based system model is modelled by incorporating the ocean current velocities into the state vector. Furthermore, to enhance system robustness under partial beam failure, a DVL fault detection and fault-tolerant strategy based on Mahalanobis distance and virtual beam reconstruction is developed to handle DVL partial beam failures. Simulation and sea experiments results demonstrate that the proposed method ensures robust navigation performance in both bottom tracking (BT) and water tracking (WT) modes. In WT mode, the positioning accuracy is improved by 66.85% compared to the LI-OC, and in BT mode, the improvement is 70.59% and 64.84% compared to LI-OC and TI-VB, respectively.

This paper focuses on the issue of interference in the strapdown inertial navigation system (SINS) and Doppler velocity log (DVL) integrated navigation system, which results in nonlinearity in underwater complex situations. Lie group theory is used to put the SINS attitude and velocity state variables into the group to change the error state. Furthermore, under the premise that the state model fails to meet the affine condition, an error state space model is reconstructed with the purpose of enhancing the velocity error equation in order to satisfy the affine condition. By deducing the nonlinear error on the lie group using a property of the linear state space model that is independent of the state estimate, filter convergence can be achieved. The simulation results indicate that the new error state model can better handle spatial inconsistency and make SINS/DVL integrated navigation more accurate and stable in nonlinear situations compared to traditional methods.