Ray-tracing method for fields of view simulation in agricultural and forestry vehicles

Fatal injuries represent a significant issue in activities involving agricultural machinery. The operator's low visibility is one of the main factors leading to such events. In this paper, a virtual prediction method of the field of view for a tractor is analyzed using a rendering based on the Ray-Tracing algorithm. Its performance is compared with standardized experimental tests based on ISO 5006:2017, presented as the "mirror test" and the "shadows test". This novel method requires the use of a three-dimensional CAD model of the vehicle under investigation, as well as the test surfaces suggested by current standards. The accuracy of the produced simulations is evaluated using several metrics, such as the amplitude, amount, and position of the masking effects. The results show that the proposed method is consistent with physical machine tests, performing better than the mirror test in all cases. Small discrepancies are due to the difficulty in synchronizing the experimental setup with the virtual model. The system accurately estimates masking effects, with an average error of 8.69% when comparing the Ray-Tracing test with the shadows test, and 26.94% with the mirror test. After improvements, the error was reduced to 6.45%.