Mercurial > hg > Members > atton > intelligence_robotics

comparison slide.md @ 10:62f384a20c2c

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
fix
author tatsuki
date Fri, 26 Jun 2015 09:09:58 +0900
parents 8a6f547b72c0
children 7104d522d2f0
comparison
equal deleted inserted replaced
9:8a6f547b72c0 10:62f384a20c2c
162 # TODO: Please Add slides over chapter (3. implementation of ARMAR-IIIb) 162 # TODO: Please Add slides over chapter (3. implementation of ARMAR-IIIb)
163 163
164 # Implementation on ARMAR-IIIb 164 # Implementation on ARMAR-IIIb
165 * ARMAR-III is designed for close cooperation with humans 165 * ARMAR-III is designed for close cooperation with humans
166 * ARMAR-III has a humanlike appearance 166 * ARMAR-III has a humanlike appearance
167 * sensory capabilities similar to humans( 167 * sensory capabilities similar to humans
168 * ARMAR-IIIb is a slightly modified version with different shape to the head, the trunk, and the hands 168 * ARMAR-IIIb is a slightly modified version with different shape to the head, the trunk, and the hands
169 169
170 # Implementation of gestures 170 # Implementation of gestures
171 * The implementation on the robot of the set of gestures it is not strictly hardwired to the specific hardware 171 * The implementation on the robot of the set of gestures it is not strictly hardwired to the specific hardware
172 * manually defining the patterns of the gestures 172 * manually defining the patterns of the gestures
175 # Master Motor Map 175 # Master Motor Map
176 * The MMM is a reference 3D kinematic model 176 * The MMM is a reference 3D kinematic model
177 * providing a unified representation of various human motion capture systems, action recognition systems, imitation systems, visualization modules 177 * providing a unified representation of various human motion capture systems, action recognition systems, imitation systems, visualization modules
178 * This representation can be subsequently converted to other representations, such as action recognizers, 3D visualization, or implementation into different robots 178 * This representation can be subsequently converted to other representations, such as action recognizers, 3D visualization, or implementation into different robots
179 * The MMM is intended to become a common standard in the robotics community 179 * The MMM is intended to become a common standard in the robotics community
180
181 # Master Motor Map
180 <img src="pictures/MMM.png" style='width: 60%; margin-left: 150px; margin-top: -50px;'> 182 <img src="pictures/MMM.png" style='width: 60%; margin-left: 150px; margin-top: -50px;'>
181 183
182 # Master Motor Map2 184 # Master Motor Map
183 * The body model of MMM model can be seen in the left-hand illustration in Figure 185 * The body model of MMM model can be seen in the left-hand illustration in Figure
184 * It contains some joints, such as the clavicula, which are usually not implemented in humanoid robots 186 * It contains some joints, such as the clavicula, which are usually not implemented in humanoid robots
185 * A conversion module is necessary to perform a transformation between this kinematic model and ARMAR-IIIb kinematic model 187 * A conversion module is necessary to perform a transformation between this kinematic model and ARMAR-IIIb kinematic model
188
189 # Master Motor Map
186 <img src="pictures/MMMModel.png" style='width: 60%; margin-left: 150px; margin-top: -50px;'> 190 <img src="pictures/MMMModel.png" style='width: 60%; margin-left: 150px; margin-top: -50px;'>
187 191
188 # converter 192 # converter
189 * converter given joint angles would consist in a one-to-one mapping between an observed human subject and the robot 193 * converter given joint angles would consist in a one-to-one mapping between an observed human subject and the robot
190 * differences in the kinematic structures of a human and the robot one-to-one mapping can hardly show acceptable results in terms of a human like appearance of the reproduced movement 194 * convert is addressed by applying a post-processing procedure in joint angle space
191 * this problem is addressed by applying a post-processing procedure in joint angle space
192 * the joint angles, given in the MMM format,are optimized concerning the tool centre point position 195 * the joint angles, given in the MMM format,are optimized concerning the tool centre point position
193 * solution is estimated by using the joint configuration of the MMM model on the robot 196 * solution is estimated by using the joint configuration of the MMM model on the robot
194 197
195 # MMM support 198 # MMM support
196 * The MMM framework has a high support for every kind of human-like robot 199 * The MMM framework has a high support for every kind of human-like robot
199 * may not be able to convert from MMM model for a specific robot 202 * may not be able to convert from MMM model for a specific robot
200 * the motion representation parts of the MMM can be used nevertheless 203 * the motion representation parts of the MMM can be used nevertheless
201 204
202 # Conversion example of MMM 205 # Conversion example of MMM
203 * After programming the postures directly on the MMM model they were processed by the converter 206 * After programming the postures directly on the MMM model they were processed by the converter
204 * Conversion is not easy
205 * the human model contains many joints, which are not present in the robot configuration 207 * the human model contains many joints, which are not present in the robot configuration
206 * ARMAR is not bending the body when performing a bow 208 * ARMAR is not bending the body when performing a bow
207 * It was expressed using a portion present in the robot (e.g., the neck) 209 * It was expressed using a portion present in the robot (e.g., the neck)
208 210
209 211
215 217
216 # Modular Controller Architecture, a modular software framework 218 # Modular Controller Architecture, a modular software framework
217 * The postures could be triggered from the MCA (Modular Controller Architecture, a modular software framework)interface, where the greetings model was also implemented 219 * The postures could be triggered from the MCA (Modular Controller Architecture, a modular software framework)interface, where the greetings model was also implemented
218 * the list of postures is on the left together with the option 220 * the list of postures is on the left together with the option
219 * When that option is activated, it is possible to select the context parameters through the radio buttons on the right 221 * When that option is activated, it is possible to select the context parameters through the radio buttons on the right
222
223 # Modular Controller Architecture, a modular software framework
220 <img src="pictures/MCA.png" style='width: 60%; margin-left: 150px; margin-top: -50px;'> 224 <img src="pictures/MCA.png" style='width: 60%; margin-left: 150px; margin-top: -50px;'>
221 225
222 # Implementation of words 226 # Implementation of words
223 * Word of greeting uses two of the Japanese and German 227 * Word of greeting uses two of the Japanese and German
224 * For example,Japan it is common to use a specific greeting in the workplace 「otsukaresama desu」 228 * For example,Japan it is common to use a specific greeting in the workplace 「otsukaresama desu」
235 * These words have been　recorded through free text-to-speech software into wave files　that could be played by the　robot 239 * These words have been　recorded through free text-to-speech software into wave files　that could be played by the　robot
236 * ARMAR does not have embedded speakers in its body 240 * ARMAR does not have embedded speakers in its body
237 * added two small speakers behind the head and connected them to another computer 241 * added two small speakers behind the head and connected them to another computer
238 242
239 # Experiment description 243 # Experiment description
240 * Experiments were conducted at room as shown in Figure 9 , Germany 244 * Experiments were conducted at room as shown in Figure , Germany
241 <img src="pictures/room.png" style='width: 60%; margin-left: 150px; margin-top: -50px;'> 245 <img src="pictures/room.png" style='width: 60%; margin-left: 150px; margin-top: 50px;'>
242 246
243 247
244 # Experiment description2 248 # Experiment description2
245 * Participants were 18 German people of different ages, genders, workplaces 249 * Participants were 18 German people of different ages, genders, workplaces
246 * robot could be trained with various combinations of context 250 * robot could be trained with various combinations of context
266 * The number of interactions taking repetitions into account was 30 270 * The number of interactions taking repetitions into account was 30
267 1. gender :M: 18; F: 12 271 1. gender :M: 18; F: 12
268 2. average age: 29.43 272 2. average age: 29.43
269 3. age standard deviation: 12.46 273 3. age standard deviation: 12.46
270 274
271 # The purpose of the experiment
272 * The objective of the experiment was to adapt ARMAR-IIIb greeting behaviour from Japanese to German culture
273 * the algorithm working for ARMAR was trained with only Japanese sociology data and two mappings M0J were built for gestures and words
274 * After interacting with German people, the resulting mappings M1 were expected to synthesize the rules of greeting interaction in Germany
275
276 # The experiment protocol is as follows 1~5 275 # The experiment protocol is as follows 1~5
277 1. ARMAR-IIIb is trained with Japanese data 276 1. ARMAR-IIIb is trained with Japanese data
278 2. encounter are given as inputs to the algorithm and the robot is prepared 277 2. encounter are given as inputs to the algorithm and the robot is prepared
279 3. Participants entered the room , you are prompted to interact with consideration robot the current situation 278 3. Participants entered the room , you are prompted to interact with consideration robot the current situation
280 4. The participant enters the room 279 4. The participant enters the room
290 # Results 289 # Results
291 * It referred to how the change in the gesture of the experiment 290 * It referred to how the change in the gesture of the experiment
292 * It has become common Bowing is greatly reduced handshake 291 * It has become common Bowing is greatly reduced handshake
293 * It has appeared hug that does not exist in Japan of mapping 292 * It has appeared hug that does not exist in Japan of mapping
294 * This is because the participants issued a feedback that hug is appropriate 293 * This is because the participants issued a feedback that hug is appropriate
294
295 # Results
295 <img src="pictures/GestureTable.png" style='width: 60%; margin-left: 150px; margin-top: -50px;'> 296 <img src="pictures/GestureTable.png" style='width: 60%; margin-left: 150px; margin-top: -50px;'>
296 297
297 # Results2 298 # Results
298 * The biggest change in the words of the mapping , are gone workplace of greeting 299 * The biggest change in the words of the mapping , are gone workplace of greeting
299 * Is the use of informal greeting as a small amount of change 300 * Is the use of informal greeting as a small amount of change
300 * some other patterns can be found in the gestures mappings judging from the columns in Table 3 for T = 0 301
301 * Japan there is a pattern to the gesture by social distance 302 # Results
302 * But in Germany not the pattern
303 * This is characteristic of Japanese society
304 * The two mapping has been referring to the feedback of the Japanese sociology literature and the German participants
305 <img src="pictures/GreetingWordTable.png" style='width: 60%; margin-left: 150px; margin-top: -50px;'> 303 <img src="pictures/GreetingWordTable.png" style='width: 60%; margin-left: 150px; margin-top: -50px;'>
306 304
307 # Limitations and improvements 305 # Limitations and improvements
308 * In the current implementation, there are also a few limitations
309 * The first obvious limitation is related to the manual input of context data 306 * The first obvious limitation is related to the manual input of context data
310 * →　 The integrated use of cameras would make it possible to determine features such as gender, age, and race of the human 307 * The integrated use of cameras would make it possible to determine features such as gender, age, and race of the human
308
309 # Limitations and improvements
310 * Robot itself , to determine whether the greeting was correct
311 * Speech recognition system and cameras could also detect the human own greeting 311 * Speech recognition system and cameras could also detect the human own greeting
312 * → Robot itself , to determine whether the greeting was correct
313 * The decision to check the distance to the partner , the timing of the greeting , head orientation , or to use other information , whether the response to a greeting is correct and what is expected 312 * The decision to check the distance to the partner , the timing of the greeting , head orientation , or to use other information , whether the response to a greeting is correct and what is expected
314 * Accurate der than this information is information collected using a questionnaire 313
314 #Limitations and improvements
315 * It is possible to extend the set of context by using a plurality of documents 315 * It is possible to extend the set of context by using a plurality of documents
316 * This in simplification of greeting model was canceled
317 316
318 # Different kinds of embodiment 317 # Different kinds of embodiment
319 * Humanoid robot has a body similar to the human 318 * Humanoid robot has a body similar to the human
320 * robot can change shape , the size capability 319 * robot can change shape , the size capability
321 * Type of greeting me to select the appropriate effect for each robot 320 * Type of greeting me to select the appropriate effect for each robot