🚀 go-pugleaf

I'm going to be writing/implementing an interpreted language to control
our underwater robot.  I've taken one course in compilers, and know (more
or less) lex and yacc (enough to have implemented a pascal subset once..
well actually a multifunction calculator that understood pascal but didn't
do anything except math functions.)

The parser will be running under the VxWorks real-time OS, and will be
listening at a TCP socket for commands.

So I'm looking for suggestions.  Some desired features of the language:
variable manipulation
math functions
run commands "in background" ala csh (spin off processes)
send "goto xyzb" commands to controller
set controller parameters
"read" from sensors, "write" to effectors
run "compiled" version of language from file
"install" sensor or effector "device driver" at an address
other things??

I've always had warm fuzzy feelings about postfix languages (forth, postscript)
but have been told that they are bad because they are "write once"; i.e. hard
to modify to change the behavior.  Perhaps a lispish language?  Maybe I should
just grab the csh parser and modify it?

Any advice would be appreciated.


-Roger
-- 
"The question of whether a computer can think is no more interesting
 than the question of whether a submarine can swim" - Edsgar W. Dijkstra 
rg@[msel|unhd].unh.edu        |  UNH Marine Systems Engineering Laboratory
r_gonzalez@unhh.bitnet        |  Durham, NH  03824-3525