Partial shading in photovoltaic (PV) arrays renders conventional maximum power point tracking (MPPT) techniques ineffective. The reduced efficiency of shaded PV arrays is a significant obstacle in the rapid growth of the solar power systems. Thus, addressing the output power mismatch and partial shading effects is of paramount value. Extracting the maximum power of partially shaded PV arrays has been widely investigated in the literature. The proposed solutions can be categorized into four main groups. The first group includes modified MPPT techniques that properly detect the global MPP. They include power curve slope, load-line MPPT, dividing rectangles techniques, the power increment technique, instantaneous operating power optimization, Fibonacci search, neural networks, and particle swarm optimization. The second category includes different array configurations for interconnecting PV modules, namely series-parallel, total-cross-tie, and bridge-link configurations. The third category includes different PV system architectures, namely centralized architecture, series-connected microconverters, parallel-connected microconverters, and microinverters. The fourth category includes different converter topologies, namely multilevel converters, voltage injection circuits, generation control circuits, module-integrated converters, and multiple-input converters. This paper surveys the proposed approaches in each category and provides a brief discussion of their characteristics.