In practical missions, quadrotor drones frequently face the challenge of navigating through multiple predetermined waypoints in cluttered environments where the sequence of the waypoints is not specified. This study presents a comprehensive multi-waypoint motion planning framework for quadrotor drones, comprising multi-waypoint trajectory planning and waypoint sequencing. To generate a trajectory that follows a specified sequence of waypoints, we integrate uniform B-spline curves with a bidirectional A* search to produce a safe, kinodynamically feasible initial trajectory. Subsequently, we model the optimization problem as a quadratically constrained quadratic program (QCQP) to enhance the trackability of the trajectory. Throughout this process, a replanning strategy is designed to ensure the traversal of multiple waypoints. To accurately determine the shortest flight time waypoint sequence, the fast marching (FM) method is utilized to efficiently establish the cost matrix between waypoints, ensuring consistency with the constraints and objectives of the planning method. Ant colony optimization (ACO) is then employed to solve this variant of the traveling salesman problem (TSP), yielding the sequence with the lowest temporal cost. The framework’s performance was validated in various complex simulated environments, demonstrating its efficacy as a robust solution for autonomous quadrotor drone navigation.

Accurate State of Charge (SoC) estimation is pivotal in advancing battery technology. In order to enhance the precision of SoC estimation, this study introduces the 2RC equivalent circuit model for lithium batteries. The Adaptive Extended Sliding Innovation Filter (AESIF) algorithm merges the model’s predictive outcomes with observation results. However, further improvements are required for this algorithm to perform optimally in strong noise environments. By adapting to observation noise and utilizing PID control to adjust the sliding boundary layer, the algorithm can accommodate varying noise levels and control interference fluctuations within specific limits. This study enhances the AESIF algorithm in these areas, proposing an improved version (IAESIF) to elevate performance in strong noise environments and improve overall estimation accuracy. Comprehensive tests were conducted under diverse operational conditions and temperatures, with results indicating that, compared to the EKF and the AESIF algorithm in strong noise environments, the IAESIF algorithm demonstrates improved noise adaptation and overall estimation accuracy.

Soft errors pose a critical reliability threat to deep neural networks in safety-critical systems. We present YOLO-FI, a fault injection framework for YOLO models that supports module-wise analysis, multiple formats, and customizable fault models. Using YOLO-FI, we evaluate YOLOv8 in floating- and fixed-point forms on COCO. Results show that floating-point YOLOv8 is vulnerable across modules, while fixed-point models are resilient except for the final module. Range restriction improves resilience in most floating-point modules but fails in the final one. To address this, we propose selective hardening of the vulnerable module, achieving effective fault mitigation with modest overhead.