The Malaysian palm oil industry has seen increased adoption of large-terrain three-dimensional (3D) mapping for assessing tree characteristics and plantation health. Nevertheless, the high computational costs and lack of auxiliary data in 3D Light Detection and Ranging (LiDAR) mapping systems impose barriers for smaller robotic companies and applications requiring smaller-sized, power-efficient robots. To overcome these issues, a cost-effective Simultaneous Localization and Mapping (SLAM) system is developed, which combines a 3D LiDAR and a depth camera, utilizing the R3LIVE framework to integrate RGB data for enhanced map visualization. The system employs an Unscented Kalman Filter (UKF) and a voxel filter to reduce noise and improve accuracy. Results indicate that the proposed SLAM system significantly reduces noise in Inertial Measurement Units (IMU) data and accurately tracks robot positioning.