 Hello everyone. I'm Leon Ampecli from the Birobots Club at Queens University in Canada. In this talk, we are going to focus on modeling HAPT simulation systems with high order models. A HAPT system usually samples position and provides force as feedback to the user in accordance with the virtual environment implemented. A typical model is shown here where the HAPT device is represented as a second order dynamic model and it's assuming an ideal relationship between the VE and activated torque. One issue is the sampling and holding process of digital platforms that restricts the VE dynamic range, but studies have shown that increasing sampling frequency is a relevant way to maximize the stable range. However, as can be seen on this plot of stiffness versus sampling frequency, there is a mismatch between the ideal model and the experimental results conducted in a one-off platform. Therefore, we propose high order models where a DC model powered by a voltage driver is used. The force model is called HOF. Here we consider the inductance and the back-MF effect with compensation in discrete. The second model, HOI, is a simplified model where it's considering that the back-MF is fully compensated and therefore only the inductance is present in this model. Here we have the uncoupled stability numerical results for comparing stiffness versus sampling frequency. We can see that the proposed models show a different relationship compared to the ideal model. For high sampling frequencies, both models show that maximum stiffness is less than provided by the ideal model. In addition, for low sampling frequencies, we can see an apparent increase in stiffness due to the back-MF compensation error. For fidelity analysis, we analyze the magnitude response of each model and compare it to the ideal model. We can see a significant difference between the models. Therefore, we could conclude that while more inductance affects the V-range, the residual back-MF mainly affects the system fidelity. For experimental analysis, we use the QEC platform with a high-resolution encoder. Here we have the uncoupled stability results for stiffness versus sampling frequency. We can see that the proposed model provides less error predicting the maximum stiffness, especially for high sampling frequencies. Additional analysis is presented in the paper. In this way, the proposed models highlighted that while inductance mainly affects the stable region, the residual back-MF degradates the system fidelity. In the future, we aim to analyze the system when a current driver is used. Thank you.