Mercurial > cortex

comparison thesis/cortex.org @ 461:b345650a0baa

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
complete first draft of final chapter.
author Robert McIntyre <rlm@mit.edu>
date Thu, 27 Mar 2014 18:17:23 -0400
parents 26c13c42481f
children bb81cef09ad7
comparison
equal deleted inserted replaced
460:763d13f77e03 461:b345650a0baa
1133 completely scrambled. The cross shape is just for convienence. This 1133 completely scrambled. The cross shape is just for convienence. This
1134 example justifies the use of pre-defined touch regions in =EMPATH=. 1134 example justifies the use of pre-defined touch regions in =EMPATH=.
1135 1135
1136 * Contributions 1136 * Contributions
1137 1137
1138 I created =CORTEX=, a complete environment for creating simulated 1138 In this thesis you have seen the =CORTEX= system, a complete
1139 creatures. Creatures can use biologically inspired senses including 1139 environment for creating simulated creatures. You have seen how to
1140 touch, proprioception, hearing, vision, and muscle tension. Each 1140 implement five senses including touch, proprioception, hearing,
1141 sense has a uniform API that is well documented. =CORTEX= comes with 1141 vision, and muscle tension. You have seen how to create new creatues
1142 multiple example creatures and a large test suite. You can create 1142 using blender, a 3D modeling tool. I hope that =CORTEX= will be
1143 new creatures using blender, a free 3D modeling tool. I hope that 1143 useful in further research projects. To this end I have included the
1144 =CORTEX= will prove useful for research ranging from distributed 1144 full source to =CORTEX= along with a large suite of tests and
1145 swarm creature simulation to further research in sensory 1145 examples. I have also created a user guide for =CORTEX= which is
1146 integration. 1146 inculded in an appendix to this thesis.
1147 1147
1148 1148 You have also seen how I used =CORTEX= as a platform to attach the
1149 /action recognition/ problem, which is the problem of recognizing
1150 actions in video. You saw a simple system called =EMPATH= which
1151 ientifies actions by first describing actions in a body-centerd,
1152 rich sense language, then infering a full range of sensory
1153 experience from limited data using previous experience gained from
1154 free play.
1155
1156 As a minor digression, you also saw how I used =CORTEX= to enable a
1157 tiny worm to discover the topology of its skin simply by rolling on
1158 the ground.
1159
1160 In conclusion, the main contributions of this thesis are:
1161
1162 - =CORTEX=, a system for creating simulated creatures with rich
1163 senses.
1164 - =EMPATH=, a program for recognizing actions by imagining sensory
1165 experience.
1149 1166
1150 # An anatomical joke: 1167 # An anatomical joke:
1151 # - Training 1168 # - Training
1152 # - Skeletal imitation 1169 # - Skeletal imitation
1153 # - Sensory fleshing-out 1170 # - Sensory fleshing-out