Automated antenna design has already become the latest trend of antenna design. An approach, which is based on the genetic algorithm (GA) and the NEC2 (numerical electromagnetics code, version 2) in conjunction with the parallel computation technique, is explored in this paper to design and optimize helical antennas. Three kinds of antennas have been optimized by the approach: a uniform helical antenna with a gain of 12.68dB, which is 3.38dB higher than that created by traditional design methods; a tapered helical antenna with average gain 12.28dB in frequency band 400~800MHz; and a tapered helix-conical horn antenna. Results show the approach is more efficient than traditional design methods and can solve different antema design problems of widely varying complexity, including those too d