Algorithm 2. Pseudo code for turning features extraction.

01: #Detected Turns     02: Access turn start and end timings with turning angle magnitude     03: #Spatiotemporal turning characteristics     04: for total turns detected     05:  Turn time = mean of (turn end time − turn start time)     06:  Turn angle = mean of turning angle magnitude vector     07:  minimum, maximum, and variability in turn time and angle vector     08:  #Full turn angular velocity     09:  Angular velocity = Turn angle/Turn time     10:   #Angular frequency (angular velocity) in start, mid & end of turn     11:  selection of 0.1 s in the start, mid, end of within turn     12:  mean & variability in (max of angular frequency in overall turn)     13:  mean & variability in (mean of 0.1 s window in the start, mid, end)     14:  #Direction of turn     15:  left turn if angle magnitude is negative or right otherwise     16:  #Number of turn/transitions     17:  length of start and end time vector     18: end for     19: #Signal based turning characteristics     20: Accessing the ω, a in VT, AP & ML directions     21: Detrend the ω, a     22: Data filtering: 4th order low-pass Butterworth filter at 20 Hz cut-off     23: Getting resultant magnitude of VT,AP & ML for ω, a     24: for each start, mid, and end of turn     25:  RMS for ω, a in VT, AP, ML & R #Turn overall, start, mid & end     26:  #Jerk: rate of change of a     27:  RMS, max, min, range of each turn jerk in VT, AP, ML & R #Turn     28:   overall, start, mid & End     29:  #Angular acceleration: rate of change of ω     30:  RMS, max, min, range of each turn angular acceleration in VT, AP,     31:   ML & R #Turn overall, start, mid & end     32: end for #classification modeling and statistical analysis