|
| Author/Centre | Blasco et al., 2002. Instituto Valenciano de Investigaciones Agrarias (IVIA), Spain [59] |
| Application | Non-chemical weed controller for vegetable crops |
| Sensorial System | Two machine vision systems: one in front of the robot for weed detection; the other for correcting inertial perturbations |
| Results | The system was able to eliminate 100% of small weeds. The system properly located 84% of weeds and 99% of lettuces |
|
| Author/Centre | Lee et al., 1999. Biological and Agricultural Engineering, University of California, USA [60] |
| Application | Real-time intelligent robotic weed control system for selective herbicide application to in-row weeds |
| Sensorial System | Two machine vision systems: one in front of the robot for guidance; the other for weed detection |
| Results | 24.2% of the tomatoes were incorrectly identified and sprayed, and 52.4% of the weeds were not sprayed |
|
| Author/Centre | Leemans and Destain, 2007. Gembloux Agricultural University, Belgium [61] |
| Application | Positioning seed drills relative to the previous lines while sowing |
| Sensorial System | Machine vision for guidance |
| Results | The standard deviation of the error was 23 mm, with a range of less than 100 mm |
|
| Author/Centre | Pérez-Ruiz et al., 2012. University of California, Davis, Department of Biological and Agricultural Engineering, USA [8] |
| Application | Automatic mechanical intra-row weed control for transplanted row crops |
| Sensorial System | RTK-GPS for controlling the path of a pair of intra-row weed knives |
| Results | A mean error of 0.8 cm in centering the actual uncultivated close-to-crop zone about the tomato main stems, with standard deviations of 1.75 and 3.28 cm at speeds of 0.8 and 1.6 km/h, respectively |
|
