Conventional pneumatic bending actuators cannot perform the in-grasp twisting required to detach mature Agaricus bisporus from the cultivation substrate. Manual harvesting was characterized by measuring fingertip contact forces, fruiting-body rotation, and motion timing using flexible-film force sensors and an inertial measurement unit. The resulting grasp–twist–lift characteristics informed the design of a three-finger segmented variable-angle symmetric oblique-chamber (SVA-SOC) pneu-net soft gripper. Each SVA-SOC actuator integrates upper parallel chambers and lower symmetric oblique chambers that are synchronously pressurized to generate inward bending for grasping; after grasp establishment, differential pressure between the lower oblique chamber groups induces torsional deformation while the grasping pressure is maintained. A reduced-order model based on the Yeoh constitutive model and segmented constant-curvature assumption was validated against finite-element simulations and experiments, yielding root mean square error (RMSE) values of 12.82° and 3.36° for bending and twisting angles, respectively. At 85 kPa, a single actuator generated bending and tangential contact forces of 1.41 and 1.31 N, respectively. The three-finger gripper achieved a maximum target rotation of 44.59°, an initial grasping force of 5.38 N, and a pull-off force of 18.59 N. In 50 harvesting trials, grasping, harvesting, and undamaged harvesting rates were 100%, 94% and 86%, respectively, indicating feasibility for low-damage harvesting of mature Agaricus bisporus.

To address the challenges of small fastener targets, complex backgrounds, and the low efficiency of traditional manual inspection in coating workshop scenarios, this paper proposes an improved Cascade R-CNN-based fastener detection method. A VOC-format dataset was constructed covering three target categories—Marking-painted fastener, Fastener, and Fallen off—which represents typical inspection scenarios of coating equipment under diverse operating conditions and enhances the adaptability of the model. Within the Cascade R-CNN framework, three improvements were introduced: the Convolutional Block Attention Module (CBAM) was integrated into the ResNet-101 backbone to enhance feature representation of small objects; anchor scales were reduced to better align with the actual size distribution of fasteners; and Soft-NMS was adopted in place of conventional NMS to effectively reduce missed detections in overlapping regions. Experimental results demonstrate that the proposed method achieves a mean Average Precision (mAP) of 96.60% on the self-constructed dataset, with both Precision and Recall exceeding 95%, significantly outperforming Faster R-CNN and the original Cascade R-CNN. The method enables accurate detection and missing-state recognition of fasteners in complex backgrounds and small-object scenarios, providing reliable technical support for the automation and intelligence of printing equipment inspection.