A challenge with how the body processes sensory information is one of the most important behavioral signs of autistic people. This is usually manifested as being either overly sensitive or underly sensitive to touch. To address this anomaly, many researchers have proposed wearable sensor-based systems and applications in the field of virtual environments but have neglected to conduct a proper evaluation and proof of concept for autistic people.

IMUs have gained popularity for tracking joint kinematics due to their portability and versatility. However, challenges such as limited accuracy, lack of real-time data analysis, and complex sensor-to-segment calibration procedures have hindered their widespread use. To address these limitations, we developed a portable system that integrates four IMUs to collect treadmill walking data, with ground truth values obtained from a Motion Capture System.

This dataset is associated with the manuscript entitled "Data-efficient Human Walking Speed Intent Inference". The data represent the measurements taken from 15 able-bodied human subjects as the made speed changes while walking on a treadmill. Each subject is associated with a .mat file that contains 8 variables. Four variables are associated with the training dataset while four are associated with the experimental testing protocol.

This study examined the effectiveness of a detection-based lie detection method that determines lying conditions based on facial autonomic reactions. This technique combines with two other lie detection techniques using a multi sensor fusion technique that is used in the polygraph test to differentiate moments of participants lying and telling the truth about a picked-up card from a deck of cards. Experiments were conducted with 19 participants sitting in front of a camera connected to Galvanic Skin Response (GSR) probes and ECG probes for a polygraph test.