The Earth's axial tilt is 23.44 deg. Cos(23.44 deg) is 0.9175 so worst case you only loose fractionally over 8% of the effective irradiance twice a year with a fixed elevation panel, with average (over the year) losses of fractionally over 4%. It isn't worth the extra mechanical complexity to add elevation tracking for photovoltaic or solar thermal panels. OTOH if you are driving a large mirror array for a solar furnace, elevation tracking is absolutely critical.
As long as the panel is pivoted about an equatorial axis, tracking the sun each day only requires a fixed angular speed drive with provision to reset it to its starting position overnight, however, the mechanical complexity of such a system is considerable, and unless you have sufficient economies of scale that a shaft drive linking multiple panels is practical, its going to be cheaper to use commodity satellite dish actuators with a MCU based controller that is programmed for constant speed motion during the day, reversing to its inital position overnight.