Proprioception is defined as the ability to sense limb and body position and motion. Proprioceptive signals originating from the mechanoreceptors around the ankle joint are critical for maintaining postural control and gait. Previous studies identified the detrimental effect of ankle somatosensory dysfunction on sensorimotor function such as balance control. However, a direct measure of ankle proprioception as a marker of proprioceptive dysfunction has not been identified. Recent advances in robotic rehabilitation devices allow precise movement of body segments while maintaining tight control over the exposed joint velocity. Here, we aim to design an ankle robot with 1-degree-of-freedom (DOF) – dorsiflexion/plantarflexion (DFPF) to objectively measure ankle joint proprioception in two aspects - joint position sense, and motion sense. To establish the feasibility of the device to evaluate ankle proprioceptive function, a small sample of healthy adult participants (n = 3) was recruited. A psychophysical 2-alternative forced-choice paradigm was employed for ankle position and motion sense evaluation. The ankle robot passively moved the ankle to one of two different positions or at two different stimulus velocities. Subsequently, participants verbally indicated the farthest position or the fastest motion. Based on their responses, a psychometric function was fitted, and a just-noticeable-difference (JND) threshold was established at the 75% correct response level, which served as a measure of ankle joint proprioception. The mean JND position threshold was 0.73°± 0.01°, and the mean JND motion threshold was 0.62°/s ± 0.05°/s. Test-retest reliability tests in all three subjects yielded the reliability coefficients of rT2 − T1 = 0.974 for JND position threshold, and rT2 − T1 = 0.948 for JND motion threshold. We here demonstrate that the feasibility and reliability of this new device to assess ankle proprioceptive acuity.