Intelligent fruit-picking robots have emerged as a promising solution to labor shortages and the increasing costs of manual harvesting. This review provides a systematic and critical overview of recent advances in three core domains: (i) vision-based fruit and peduncle detection, (ii) motion planning and obstacle-aware navigation, and (iii) robotic manipulation technologies for diverse fruit types. We summarize the evolution of deep learning-based perception models, highlighting improvements in occlusion robustness, 3D localization accuracy, and real-time performance. Various planning frameworks—from classical search algorithms to optimization-driven and swarm-intelligent methods—are compared in terms of efficiency and adaptability in unstructured orchard environments. Developments in multi-DOF manipulators, soft and adaptive grippers, and end-effector control strategies are also examined. Despite these advances, critical challenges remain, including heavy dependence on large annotated datasets; sensitivity to illumination and foliage occlusion; limited generalization across fruit varieties; and the difficulty of integrating perception, planning, and manipulation into reliable field-ready systems. Finally, this review outlines emerging research trends such as lightweight multimodal networks, deformable-object manipulation, embodied intelligence, and system-level optimization, offering a forward-looking perspective for autonomous harvesting technologies.

In response to the poor detection performance of grapes in orchards caused by issues such as leaf occlusion and fruit overlap, this study proposes an improved grape detection method named YOLOv7-MCSF based on the You Only Look Once v7 (YOLOv7) framework. Firstly, the original backbone network is replaced with MobileOne to achieve a lightweight improvement of the model, thereby reducing the number of parameters. In addition, a Channel Attention (CA) module was added to the neck network to reduce interference from the orchard background and to accelerate the inference speed. Secondly, the SPPFCSPC pyramid pooling is embedded to enhance the speed of image feature fusion while maintaining a consistent receptive field. Finally, the Focal-EIoU loss function is employed to optimize the regression prediction boxes, accelerating their convergence and improving regression accuracy. The experimental results indicate that, compared to the original YOLOv7 model, the YOLOv7-MCSF model achieves a 26.9% reduction in weight, an increase in frame rate of 21.57 f/s, and improvements in precision, recall, and mAP of 2.4%, 1.8%, and 3.5%, respectively. The improved model can efficiently and in real-time identify grape clusters, providing technical support for the deployment of mobile devices and embedded grape detection systems in orchard environments.

Strawberry detection in complex orchard environments remains a challenging task due to frequent leaf occlusion, fruit overlap, and illumination variability. To address these challenges, this study presents an improved lightweight detection framework, DS-YOLO, based on YOLOv8n. First, the backbone network of YOLOv8n is replaced with the lightweight StarNet to reduce the number of parameters while preserving the model’s feature representation capability. Second, the Conv and C2f modules in the Neck section are replaced with SlimNeck’s GSConv (hybrid convolution module) and VoVGSCSP (cross-stage partial network) modules, which effectively enhance detection performance and reduce computational burden. Finally, the original CIoU loss function is substituted with WIoUv3 to improve bounding box regression accuracy and overall detection performance. To validate the effectiveness of the proposed improvements, comparative experiments were conducted with six mainstream object detection models, four backbone networks, and five different loss functions. Experimental results demonstrate that the DS-YOLO achieves a 1.7 percentage point increase in mAP50, a 1.5 percentage point improvement in recall, and precision improvement of 1.3 percentage points. In terms of computational efficiency, the number of parameters is reduced from 3.2M to 1.8M, and computational cost decreases from 8.1G to 4.9G, corresponding to reductions of 43% and 40%, respectively. The improved DS-YOLO model enables real-time and accurate detection of strawberry fruits in complex environments with a more compact network architecture, providing valuable technical support for automated strawberry detection and lightweight deployment.