Assessing the Limits of Equivalent Circuit Models and Kalman Filters for Estimating the State of Charge: Case of Agricultural Robots

The battery State of Charge (SoC) is critical information to overcome agricultural robots’ limitations related and energy management. Although several SoC estimation methods have been proposed in the literature, performance these has not validated for different chemistries mobile robot applications. Compared previous work, this paper evaluates limits using RC model Thevenin a Lithium Iron Phosphate (LFP) Sealed Lead Acid (SLA) battery. This evaluation used custom controlled indoor environment. Consequently, work assessed two ECM-based packs, low-cost sensors discharge cycles typically Finally, results indicate that suitable LFP battery; however, it achieved mean absolute error (MAE) 2.2% SLA On other hand, performed properly both chemistries, achieving MAE lower than 1%.