A comparison of approaches for citrus canopy profile generation using ultrasonic and Leddar® sensors

Tree canopy profile can be generated with a variety of sensors that include Lidar, ultrasonic sensors, radar and camera-based systems. In this study, a new kind of sensor, Leddar (R) that uses LED's to measure distances is compared with a popular ultrasonic sensor. The purpose of the study is to assess the viability of the sensors to form part of a control system design that will enable the operation of an automated shaker system. The shaker system will be able to move dynamically based on the shape of the tree canopy formed by the sensor outputs. Statistical methods were used in both indoor and outdoor environments to identify and reduce the systematic and random errors involved in the sensor measurements. The sensors provided with an average error less than 5.08 cm in the indoor environment and less than 10.16 cm in the outdoor environment. The sensor outputs showed lower error values on larger distances in the outdoor setting. A vertical and horizontal grid setup formed the basis of development of 3D canopy profiles of a grove of citrus trees. This grid setup was used to generate a point cloud from the sensor outputs to represent the tree canopy profile. Volume estimation is also performed with the Leddar (R) sensors. This study enlists an experimental approach to compare the performance of ultrasonic and Leddar sensors to 3D canopy profile generation and volume estimation.