Remote Radio Control of Insect Flight

We demonstrated the remote control of insects in free flight via an implantable radio-equipped miniature neural stimulating system. The pronotum mounted system consisted of neural stimulators, muscular stimulators, a radio transceiver-equipped microcontroller and a microbattery. Flight initiation, cessation and elevation control were accomplished through neural stimulus of the brain which elicited, suppressed or modulated wing oscillation. Turns were triggered through the direct muscular stimulus of either of the basalar muscles. We characterized the response times, success rates, and free-flight trajectories elicited by our neural control systems in remotely controlled beetles. We believe this type of technology will open the door to in-flight perturbation and recording of insect flight responses.

FIgure | (A) Tetherless flight control system (∼230 mg total) mounted on Cotinis texana (Green June Beetle) using beeswax next to a US$ 0.25 coin. A microcontroller provided potential pulses to four stimulating wire electrodes (∅125 μm) implanted into the brain, left and right basalar muscles and posterior pronotum (counter electrode). (B) Radio flight control system (∼1.3 g total) mounted on Mecynorrhina torquata using beeswax next to a US$ 0.25 coin. The system consisted of a microcontroller, a custom PCB, a dipole antenna, a microbattery and stimulating wire electrodes (∅125 μm) implanted as in Cotinis. (C) Front and (D) tilted views of dissected Cotinis beetle head showing the brain stimulator at implant site 1, optic lobe stimulator at implant site 2. The brain stimulator was implanted along the rostral-caudal midline of the head, at the center between the left and right compound eyes. Implant site 2 was at the interior edge of each compound eye. (e) Sagittal section of thorax showing the counter electrode at implant site 3 and the basalar muscle stimulator at implant site 4. (F) Cross-section of mesothorax showing the basalar muscle stimulator sites (implant site 4 on left and right sides). The basalar muscle stimulator was implanted midway between sternum and notum of mesothorax to a depth of approximately 1 cm in rostralcaudal direction on either the left or right side of the insect. The blue letters X and bars indicate implant sites and approximate implant lengths, respectively. Mecynorrhina torquata has nearly identical, scaled anatomy to Cotinis texana.

FLIghT INITIATION AND CESSATION
In C. texana, alternating positive and negative potential pulses betweenanelectrodeimplantedintothebrainandacounterelectrodeimplantedintotheposteriorpronotumoftheadultinsect reproduciblygeneratedflightinitiationandcessationwithsuccess rateof56%(N=9)infullytetheredandweaklytetheredCotinis beetles(seeSection"MaterialsandMethods"); Figure 2 , which indicates that the stimulus caused not only muscle movement coordinated with wing oscillation but also uncoordinated muscle movement associated with generalized neuraldepolarization.
We then compared three different types of electrical stimuli: alternatingnegativeandpositivepotentialpulses,positivepotential pulsesandnegativepotentialpulses (Figure 3).Positivepotentials, whetheraloneoralternatingwithnegativepulses,initiatedflight but negative potential pulses alone did not. Positive pulses and alternating positive and negative pulses were equally effective in eliciting flight: five of nine and four of nine insects initiated flightinresponsetostimulation,respectively.Dataonstimulated flightboutsinindividualC. texanaaresummarizedinTable1in SupplementaryMaterial.
FIgure | Initiation and cessation control of Cotinis texana beetle during tethered flight; (top) audio recordings of tethered beetle, (bottom) applied potential to the brain (with counter electrode inserted into posterior pronotum). The applied potential waveform is identical to Figure A, but frequency varied. As the period between pulses decreased, the beetle was incapable of fully starting or stopping wing oscillation and audio amplitudes were modulated by the stimulus frequency. Audio amplitudes were normalized using mean absolute value during normal, sustained flight recorded at each individual trial. See Movies 1, 2 and 4 in Supplementary Material for flight initiations of fully tethered, weakly tethered and fully untethered Cotinis texana, respectively.
In summary, we demonstrated a miniaturized, pronotummounted system consisting of a neural stimulator, muscular stimulators,aradio-equippedmicrocontrollerandamicrobattery capableofthecontinuousflightcontrolof1g/2cmand8g/6cm beetles in free flight. To our knowledge, this is one of the first reportsonareliable,neuro-stimulatedflightcontrolmechanismin insects.Althoughtherehavebeenpriorreportsontheinfluenceof  Table 2 in Supplementary Material. The sharp rise of audio amplitude at the beginning of oscillation was attributed to friction between elytra and wings when the wings were unfolded from the underneath of elytra. The whole audio amplitudes were normalized using mean absolute value calculated for the middle period of the flight time (2.5-3.7 s). (B) Pulse trains applied between left and right optic lobes. Number of waveforms was swept from 1 to 100 in one waveform increment when testing for the number of waveforms required to trigger flight initiation ( Table 2 in (Burrows,1996),the mechanismsandmicrosystempresentedhereofferdistinctadvantagefortheremotecontrolandstudyofinsectflight.Oneofthe majoradvantagesofourmethodisthatthestimulationmethodis surprisinglysimpleandrobust,anditimplicitlymakesuseofthe beetle'sownflightcontrolcapabilities-thebeetlepowersitsown flightandlevelstothehorizon;perturbationsareappliedwhenever aheadingorelevationchangeisrequired.Theimplantmethod described here suffers from variability in stimulus voltage from insect-to-insect;thisislikelyduetothecoarsenatureofthestimulatorandtheuseofelectricalpotentialasthecontrolledvariable(as opposedtochargedelivered).Smallerfootprintmicrofabricated electrodesshouldimprovethefirstissue,aswellasreducetheoverallpowerconsumptionofeachstimulus.Moreover,newerdesigns shouldlikelyusechargedelivery(asopposedtovoltagelevels)from microcontroller-drivencurrentsourcestoelicitresponses.    Figure 3,Table1inSupplementaryMaterial).