Robots as Avatars of Evolution Under Stress*

Experimental robo-biological  physics can bring new insights into biological evolution.  We present a development of hybrid analog/digital autonomous robots with mutable diploid dominant/recessive 6-byte genomes which exhibit pleiotropy because of multi-phenotype control by discrete genes.  The robots are capable of death, rebirth and breeding.  We map the quasi-steady state surviving local density of the robots onto a multi-dimensional abstract ``survival landscape''. We show it is generally necessary for robot survival on an externally and self-modified resource landscape to require the exchange of genes between the robots in addition to mutations, and we show via measurement of the Shannon genetic entropy that high genomic diversity and pleiotropy is essential for survival in a time and space stochastic environment. We propose that diploid gene robots with pleiotropy can act as avatars of diploid mammalian cells for novel programs of administration of drugs, and that a stochastic time and drug chemotherapy course would target cancer cells more effectively.