The battery charging method uses a pulsed current alternately taking first and second amplitudes during first and second periods (t1, t2). At least one of the parameters, amplitude and duration, of the current during the first period (t1) is automatically servo-controlled as a function of a first value of the slope (P1) of the voltage (U), measured at the terminals of the battery to be charged. The first slope value (P1) is calculated at least at the end of the first period (t1). Likewise, a second value of the slope (P2) of the voltage (U) can be calculated at least at the end of the second period (t2) and at least one of the parameters of the current during the second period (t2) can be automatically servo-controlled as a function of the second slope value (P2). The different parameters of the pulsed charging current are thus continuously servo-controlled as a function of the voltage slope to keep the slope (P) within a range comprised between 1 mV/s and 6 mV/s.