中文摘要：

为了解决移动机器人障碍物检测中的多传感器数据融合问题,在基本免疫遗传算法（BIGA）基础上,提出了一种异型细胞杂交免疫进化算法（HCHIGA）.通过选取合适异型个体,对优势基因进行逻辑互补,解决了BIGA中因交叉位随机选取而造成的早熟及收敛缓慢问题;以最小化多传感器测距数据总均方差为优化目标,实现了数据的免疫融合.函数测试结果表明,与BIGA和遗传算法（GA）相比,HCHIGA优化精度更高、速度更快、稳定性更好.多传感器数据融合结果表明,与自适应加权法及BIGA相比,HCHIGA的误差分别减小0.618％和0.443％,较好地解决了多传感器数据融合问题.

英文摘要：

To solve the multi-sensor data fusion problem in the obstacle detection of mobile robots,a heterotypic cell hybridization-based immune genetic algorithm （HCHIGA） was presented based on the basic immune genetic algorithm （BIGA）.In HCH-IGA,some appropriate heterotypic cells were selected to have a logical complementation on dominant genes,which solved the problem of prematurity and slow convergence caused by random crossion positions.Then the minimum of total mean square error with respect to the multi-sensor data was taken as the optimization objective,and the immune fusion was realized.Compared with the BIGA and genetic algorithm （GA）,the function test results showed that,the proposed HCHIGA had the characteristic of high optimization precision,quick convergence and good stability.Compared with adaptive fusion and BIGA,the fusion results of multi-sensor data showed that the fusion errors of HCGIGA decreased by 0.618％ and 0.443％ respectively,which solved the multi sensor date fusion problem well.