The overall goal of this experiment is to evaluate the odor-tracking capability of an insect used as a controller of a mobile robot. This method can address key challenges in the biomedics field, such as how to utilize insects in biomimetic mobile robot navigation. The main advantage of this technique is that we can directly use an insect as a controller of a mobile robot and evaluate its ability to find an odor source.
The implications of this technique extend toward understanding of the biological mechanisms behind other behaviors because investigators can alter the sensory-motor relationship of an onboard insect by robotic manipulation. To begin the experiment, first select an adult silkmoth aged between two and eight days. Gently remove all the scales on the mesonotum using a piece of wet tissue or cotton swab to expose the cuticle.
Paste an adhesive onto the strip of plastic at the end of the attachment. Then, paste some adhesive onto the surface of the exposed mesonotum, taking care not to touch the wing hinge or forewing tegulae. Wait for five to ten minutes until the adhesive is no longer sticky.
Bond the mesonotum of the moth to the attachment. Keep the moth tethered by anchoring the attachment to a stand. Place a piece of paper under the moth's legs to allow it to rest.
Wash the surface of a white, expanded polystyrene ball with water to remove any possible olfactory or visual cues. Then, turn on the fan that supplies air to the treadmill. The ball should float approximately two millimeters above the bottom of the cup.
Using a screw, attach the copper wire of the moth attachment to the fixture in the cockpit of the robot. Make sure that the moth is positioned with the middle leg pair in the center of the ball. Whilst keeping the height of the ball the same, adjust the vertical position of the attachment to enable the moth to walk normally on the ball.
If desired, perform sex pheromone stimulation to the tethered moth to check its walking behavior. Checking the walking behavior of a tethered moth is critical in appropriate vertical position of the moth with their irregular behavior or the behavior of optical-sensory reading. Both directly affect the resulting robot movement.
Turn on the air of a pulling-air-type wind tunnel and set the wind speed to 0.7 meters per second. Ensure that the temperature is above 20 degrees Celsius. Set the odorant at the center of the cross-wind position and 25 centimeters downwind from a mesh panel.
Turn on the micro-controller board of the robot and establish a serial connection to a computer via Bluetooth. Launch the BioSignal program interface between the computer and the robot. Click on the about device"button to confirm the connection by sending a command to the robot via the specified COM port, and then check that a message is returned from the robot.
Click on the memory erase"button to remove any previous locomotion data in the onboard flash memory. Then, press the drivemode1"button to send the default motor gains to the robot. Click on don't drive"to send the command to the robot to immobilize until the experiment starts.
Place the robot at the start position 60 centimeters downstream from the odor source. Turn on the switch to the motor driver board. Next, push the record button on the camcorder to begin video capture.
Click on the record start"button to send a start command that will initiate the robot and simultaneously begin recording of the ball rotation to the onboard flash memory. The robot should begin to move and track the odor plume. When the robot localizes the odor source or does not localize within four minutes, click on the record stop"and don't drive"buttons to send commands to stop both the robot and the recording.
Finally, push the recording button of the camcorder to stop video capture. The rotational and translational velocity of the robot can also be modified to investigate the compensatory skills of the insect. To test the effect of asymmetrical motor gains, first define the rotational gains by editing the parameters file in the text editor.
In the BioSignal software, click on setparam2"to read the edited configuration. Then, click on drivemode2"to send these manipulated gains to the robot. To manipulate the olfactory input, change the gap between the suction tube tips.
Or, invert their positions to alter the odor concentration delivered to each antenna. Finally, visual input can be modified by covering the canopy with white paper to occlude 150 degrees and 90 degrees of the horizontal and vertical visual field of the onboard moth respectively. The comparison of odor-tracking behaviors between freely-walking moths and the insect-controlled robot is shown here.
Under the same odor circumstances, both the walking moths and the robots scored success rates of 100%Though the robot exhibited broader trajectories compared to the walking moth, there was no significant difference in time to odor localization. The odor-delivery system is necessary for supplying the odorant flow to the antennae of the elevated moth. Without this system, the robot could not orient toward the odor source and continued circling until it stopped.
The effectiveness of a bilateral olfactory input for odor-tracking was evaluated by changing the position of the tube tips. Compared to the control standard wide-tube gap, robots were manipulated to have either a narrow tube gap, inverted wide-tube gap, or an inverted motor output. The robot achieved 100%success with both gap width conditions, with no significant difference in time to localization.
Conversely, inversion of the tube tips broadened trajectories and increased time to localization, although not significantly. Finally, in the inverted motor condition, the robots circled continuously, significantly lengthening the time to localization. Once mastered, the implementation of a silkmoth into the lowered cockpit can be done within ten minutes if it is performed properly.
While attempting this procedure, it is important to remember to gently handle the silkmoth throughout the operation to avoid unnecessary damage. Following this procedure, the long-time experiment can be performed in order to answer additional questions concerning the mechanisms behind adaptive behaviors, such as whether insects can learn to pilot the manipulated robotic body. After its development, this technique paved the way for researchers in the field of neuroscience and robotics to explore the mechanisms of adaptive behaviors and the application to the control of autonomous robots.
