Work Estimate: 16 hrs.
Currently, the robot recognizes a set of speech elements and a set of gesture elements, but does not have any logic to combine the two inputs. With this feature, the robot would be able to recognize inputs that have a gesture component combined with a spoken component. For instance, the robot could be configured to ignore a user if they say “Hi”, and ignore a user if they wave their hand, but recognize if the user both says “Hi” and waves their hand at the same time. This would require some rework of the speech/gesture recognition code to have the robot remember recently viewed gesture/speech input, and trigger an event if a second component is detected after the first. This should work regardless of the input order (e.g. either “Hi” followed by wave, or wave followed by “Hi”), and should have some short timeout for each input (e.g. so that “Hi” followed by a 5 minute delay, followed by a wave does not trigger the response).
Work Estimate: 16 hrs. Currently, the robot recognizes a set of speech elements and a set of gesture elements, but does not have any logic to combine the two inputs. With this feature, the robot would be able to recognize inputs that have a gesture component combined with a spoken component. For instance, the robot could be configured to ignore a user if they say “Hi”, and ignore a user if they wave their hand, but recognize if the user both says “Hi” and waves their hand at the same time. This would require some rework of the speech/gesture recognition code to have the robot remember recently viewed gesture/speech input, and trigger an event if a second component is detected after the first. This should work regardless of the input order (e.g. either “Hi” followed by wave, or wave followed by “Hi”), and should have some short timeout for each input (e.g. so that “Hi” followed by a 5 minute delay, followed by a wave does not trigger the response).