Dragon fruit detection in natural environments remains challenged by limited accuracy and deployment difficulties, primarily due to variable lighting and occlusions from branches. To enhance detection accuracy and satisfy the deployment constraints of edge devices, we propose YOLOv10n-CGD, a lightweight and efficient dragon fruit detection method designed for robotic harvesting applications. The method builds upon YOLOv10 and integrates Gated Convolution (gConv) into the C2f module, forming a novel C2f-gConv structure that effectively reduces model parameters and computational complexity. In addition, a Global Attention Mechanism (GAM) is inserted between the backbone and the feature fusion layers to enrich semantic representations and improve the detection of occluded fruits. Furthermore, the neck network integrates a Dynamic Sample (DySample) operator to enhance the spatial restoration of high-level semantic features. The experimental results demonstrate that YOLOv10n-CGD significantly improves performance while reducing model size from 5.8 MB to 4.5 MB—a 22.4% decrease. The mAP improves from 95.1% to 98.1%, with precision and recall reaching 97.1% and 95.7%, respectively. The observed improvements are statistically significant (p < 0.05). Moreover, detection speeds of 44.9 FPS and 17.2 FPS are achieved on Jetson AGX Orin and Jetson Nano, respectively, demonstrating strong real-time capabilities and suitability for deployment. In summary, YOLOv10n-CGD enables high-precision, real-time dragon fruit detection while preserving model compactness, offering robust technical support for future robotic harvesting systems and smart agricultural terminals.

This paper proposes an enhanced RRT* algorithm (XN-RRT*) to address the challenges of low path planning efficiency and suboptimal picking success rates in complex pitaya harvesting environments. The algorithm generates sampling points based on normal distribution and dynamically adjusts the center and range of the sampling distribution according to the target distance and tree density, thus reducing redundant sampling. An improved artificial potential field method is employed during tree expansion, incorporating adjustment factors and target points to refine the guidance of sampling points and overcome local optima and infeasible targets. A greedy algorithm is then used to remove redundant nodes, shorten the path, and apply cubic B-spline curves to smooth the path, improving the stability and continuity of the robotic arm. Simulations in both two-dimensional and three-dimensional environments demonstrate that the XN-RRT* algorithm performs effectively, with fewer iterations, high convergence efficiency, and superior path quality. The simulation of a six-degree-of-freedom robotic arm in a pitaya orchard environment using the ROS2 platform shows that the XN-RRT* algorithm achieves a 98% picking path planning success rate, outperforming the RRT* algorithm by 90.32%, with a 27.12% reduction in path length and a 14% increase in planning success rate. The experimental results confirm that the proposed algorithm exhibits excellent overall performance in complex harvesting environments, offering a valuable reference for robotic arm path planning.