Transmitted vibration from vehicles to driver body generates some problems in the long term. Passive and active seat suspensions are used in heavy duty vehicles to reduce unwanted vibration and prevent health problems due to oscillation. Seat suspension must minimize the driver's body displacement and acceleration to increase riding convenience. Active force control (AFC) method is a new technique which is used in active controllers and makes them more accurate. Therefore, this work represents the possibility of applying AFC strategy for an active seat suspension control to increase its robustness. An AFC-based scheme is designed and simulated in MATLAB software. In addition, artificial neural network (ANN) is integrated into the AFC loop to approximate estimated mass of the seat and human body for the proposed controller. The training of ANN with multi-layer feedforward structure is carried out using Levenberg-Marquardt learning algorithm. To evaluate the neuro-AFC control system robustness, the seat is subjected to various types of disturbances. The results of the present study illustrate that the neuro-AFC technique is computationally simple and efficient compared to the classic proportional-integral-derivative (PID) controller in suppressing undesired vibration of heavy duty vehicles' seat. The neuro-AFC scheme is found to demonstrate superior performance for various road profiles compared to pure PID controller, and it can be successfully utilized in heavy duty vehicles such as industrial and agricultural tractors.

Whole body vibration produces some serious problems for human health in the long term. Low-frequency vibration, generated during vehicle operation, and transmitted to the vehicle operator, plays a major role in the development of low-back pain. Back pain is one of epidemic injuries in heavy duty vehicle drivers. Generally seat suspensions are designed and optimised to remove this unwanted movement. Human body biodynamic model is essential in passive seat suspension optimisation and active control seat suspension design. Lumped parameter models have been used by researchers for this purpose, but they have some limitations such as fixed body weight. With reference to this limitation, in first part of this paper a new artificial neural network (ANN) model is introduced which can predict spine acceleration from excitation signal and human body mass and height. The accuracy of model is 96% and makes it useful in real-time and off-line analysis. In second part of the paper, an off-road seat suspension will be optimised via this achieved ANN model and three Meta-Heuristic algorithms.

Since one of the influential factors that affects the spray distribution pattern is the spray boom movements which are mostly induced by soil unevenness, most of the recent sprayers are equipped with suspensions for improving the uniformity of spray application in the field. This paper investigates the suitability of improving the sprayer suspension dynamics performance by employing a robust intelligent control scheme, namely active torque control (ATC) based method in reducing the undesired vibration through a simulation study. The ATC scheme with a self-tuning fuzzy proportional-integral-derivative (PID) (ATC-STF-PID) controller was first designed and simulated. Then an artificial intelligence (AI) method using iterative learning (IL) was embedded and implemented into the ATC loop to compute the estimated inertial parameter of the system; this scheme is known as ATCAIL. Thereafter, the performance of the ATCAIL scheme is later compared to the ATC with artificial neural network (ATCANN), ATC-STF-PID and STF-PID controllers in time and frequency domains. The results of simulation work affirm that ATC-based schemes can improve the system performance of the active rolling suspension in relation to roll vibration suppression. In other words, both the ATCAIL and ATCANN schemes show better responses in comparison to the ATC-STF-PID controller scheme. The results also imply that the ATCAIL scheme is indeed effective in suppressing the vibration of a sprayer boom structure.

Currently, most of modern sprayers are equipped with suspensions for improving the uniformity of spray application in the field. Therefore, this paper represents the possibility of applying active force control (AFC) technique for the control of a spray boom structure undesired roll movement through a simulation analysis. The dynamic model of the spray boom was firstly defined and an AFC-based scheme controller was designed and simulated in MATLAB environment.Artificial neural network (ANN) is incorporated into the AFC scheme to tune the proportionalderivative (PD) controller gains andcompute the spray boom estimated mass moment of inertia. The training of both ANN with multi layer feed forward structure was done using Levenberg-Marquardt (LM) learning algorithm. To evaluate the AFC-ANN control system robustness, various types of disturbances and farmland terrain profileshave been used to excite the spray boom. The results of the study demonstrated that the AFC-based method offers a simple and effective computation compared to the conventional proportional-integral-derivative (PID) control technique in attenuating the unwanted spray boom roll oscillation or vibration. The AFC-ANN scheme is found to exhibit superior performance for different proposed terrain profilesin comparison to the AFC-PD and pure PD counterparts.

Nowadays, usage of vehicles increases due to modern lifestyles, and many people are exposed to vibrations in vehicles. Vibrations in low frequency range cause some serious long-term diseases in both aspects physically and psychologically. Vibration model helps researchers to have better interpretation of vibrations transmitting to human organs. Lumped models are very popular in this field, and different types of models with various degrees of freedom have been introduced. The main disadvantage of lumped models is that due to its fixed weight, some modifications need to be made to new subjects. Therefore, a new biodynamic model with artificial neural network method was constructed to simulate transmitted vibration to head for seated human body by conducting indoor vertical vibration experiments. Five healthy males participated in the tests. They were subjected to vertical vibration, and their responses were recorded. A neural network model was trained by input-output accelerations. The developed model was able to predict head acceleration from exciting vibration at the pelvic. In addition, weight and height of human body were considered as input factors. The comparison between the model evaluation results and the experimental and other lumped models affirmed high accuracy of the achieved artificial neural network biodynamic model.