OBJECTIVE An exploratory study to identify parameters that can be used for estimating a subject's cardio-respiratory physical fitness level, expressed as VO2max, from a combination of heart rate and 3D accelerometer data. METHODS Data were gathered from 41 healthy subjects (23 male, 18 female) aged between 20 and 29 years. The measurement protocol consisted of a sub-maximal single stage treadmill walking test for VO2max estimation followed by a walking test at two different speeds (4 and 5.5 kmh-1) for parameter determination. The relation between measured heart rate and accelerometer output at different walking speeds was used to get an indication of exercise intensity and the corresponding heart rate at that intensity. Regression analysis was performed using general subject measures (age, gender, weight, length, BMI) and intercept and slope of the relation between heart rate and accelerometer output during walking as independent variables to estimate the VO2max. RESULTS A linear regression model using a combination of the slope and intercept parameters, together with gender revealed the highest percentage of explained variance (R2 = 0.90) and had a standard error of the estimate (SEE) of 2.052 mL O2kg-1min-1 with VO2max. Results are comparable with current commonly used sub-maximal laboratory tests to estimate VO2max. CONCLUSIONS The combination of heart rate and accelerometer data seems promising for ambulant estimation of VO2max-.