To address the challenges of inefficient Camellia oleifera fruits harvesting in hilly and mountainous regions due to the difficulty of using large machinery, a handheld vibration harvesting device for Camellia oleifera fruits was designed. Based on the vibration-induced detachment process of Camellia oleifera fruits, a single-pendulum dynamic model of the “fruit-branch” system was established and solved to calculate the tangential acceleration required for fruit detachment. The key factors influencing harvesting efficiency were identified as vibration frequency, amplitude, height, and duration. Using ANSYS, modal response and harmonic response analyses were conducted on a 3D model of the Camellia oleifera tree to determine the operational parameters ensuring branch acceleration meets the fruit detachment. Furthermore, a rigid-flexible coupling simulation system integrating the harvesting device and Camellia oleifera tree was developed on the ADAMS. This analysis revealed the variation patterns of branch acceleration with respect to vibration frequency and clamping height, thereby validating the rationality of the dynamic model and the feasibility of the device. Finally, an orthogonal experiment was designed using Design-Expert 13, and multi-objective optimization analysis was performed on the device’s working parameters based on the experimental data. The aforementioned research identified the optimal working parameter combination and actual harvesting performance of the handheld vibration harvesting device: when the vibration frequency is 14 Hz, vibration height is 980 mm, and vibration duration is 13 s, the fruit picking rate reaches 95.22%. The harvesting efficiency of this device is significantly higher than manual picking methods, fully meeting the requirements for efficient Camellia oleifera fruit harvesting.

High-fidelity acquisition of canopy phenotypic data is critical for the advancement of orchard Artificial Intelligence (AI). Yet, an improper Light Detection and Ranging (LiDAR) installation height (IH) frequently induces data occlusion and substantial measurement errors. To address this limitation, this study developed an information collection vehicle (ICV) integrated with a 16-channel three-dimensional (3D) LiDAR to determine the optimal LiDAR IH. Three representative LiDAR IHs (1.4 m, 2.0 m, and 2.6 m) were evaluated on spindle-shaped cherry trees under both forward and reverse driving strategies. Subsequently, a novel 12-zone refined evaluation framework was introduced to quantify localized errors that are conventionally obscured by traditional whole-canopy metrics. Results demonstrated a profound nonlinear relationship between IH and measurement accuracy. Specifically, the 2.0 m IH (approximating the canopy’s geometric center) emerged as the optimal setup, maintaining relative errors (REs) below 5% with minimal dispersion. Conversely, the 2.6 m IH caused lower-canopy volume REs to surge beyond 16% owing to restricted downward viewing angles. Additionally, reverse driving at higher IHs exacerbated mechanical vibrations via the “lever arm effect”, thereby significantly degrading point cloud registration accuracy. Ultimately, these findings underscore the critical necessity of aligning sensors with the canopy geometric center, supplying essential theoretical guidelines for the hardware design of future orchard robots.