Parkinson’s disease (PD) is characterized by progressive motor and sensorimotor deficits, including impairments in proprioception and grip force control. Despite their clinical relevance, these alterations are rarely evaluated quantitatively. In this study, we present a portable, sensor-based system designed to assess proprioceptive force perception in PD through a structured grip force protocol. Participants performed a three-phase task involving visually guided and unguided contractions at 50% of their maximum voluntary contraction (MVC). Signals were acquired at 250 Hz and processed offline to extract multiple performance metrics, including root mean squared error (RMSE), standard deviation, average percentage distance to the 50% MVC target, and the slope during the plateau region. Data were analyzed from 19 patients with PD and 32 healthy controls. Results showed that PD patients consistently applied less force than the 50% MVC target during the phase without visual feedback and exhibited greater deviations during sustained contractions (plateau region). Statistically significant group differences revealed characteristic patterns of error consistent with proprioceptive dysfunction, particularly in the average percentage distance between the applied force and the 50% MVC target during the plateau region (p < 0.02, Kolmogorov–Smirnov test). These findings support the potential of quantitative tools to identify sensorimotor alterations in PD and contribute to clinical assessment protocols.