Downloaded 987 times
![• TOTAL of 12 samples (3 per sensor)
are recorded in 12-variable vectors
• THE PCA-based feature-extraction
algorithm calculates the eigenvectors and Eigen values
based on the 12-variable vectors
• THREE eigenvectors with the largest
Eigen values are then chosen to set up the feature
matrix [v1, v2, v3]](https://image.slidesharecdn.com/tonguedrivesystembyanu-120922012417-phpapp02/85/Tongue-drive-system-21-320.jpg)
Uploaded byAnagha Balakrishnan
Tongue drive system
The document describes the Tongue Drive System (TDS), an assistive technology that allows people with severe disabilities to control their environment using only their tongue. [1] TDS uses a small magnet pierced on the tongue and magnetic sensors to detect tongue movements and send wireless signals to control devices like powered wheelchairs. [2] It provides an alternative to existing assistive technologies that have limitations. The document outlines the components, working, advantages and limitations of the TDS.
More Related Content
Similar to Tongue drive system
Tongue drive system
- 1. TONGUE DRIVE SYSTEM Presented by: Anagha Balakrishnan ER & DCI-IT,Vellayambalam RollNo: 07
- 2. What is TDS? TONGUE–OPERATED assistive technology for people with severe disability to control their environment DEVELPOED by GT Bionics Lab
- 3. Assistive Technology ? ASSISTIVETECHNOLOGY is an umbrella term that includes assistive, adaptive, and rehabilitative devices for people with disabilities and also includes the process used in selecting, locating, and using them.
- 4. Why Spinal CordInjury is Fatal? •SPINAL CORD is located inside the spine •CONNECTS the brain to nerves in the body •MESSAGE travel in very high speed from brain to rest of the body
- 5. Cause of SpinalCord Injury… SPINAL CORD PARALYSIS is squashed or blood circulation of the spinal cord is cut off, which will result in.. UNABLE to easy breath UNABLE to feel pain or sensation SORES on skin not being able to move around SEXUAL dysfunction
- 6. Existing Assistive Technologies… Sip-and-Puff wheel chair •USE air pressure to control wheel chair by sipping and puffing •NOT good for people with week breathing
- 7. Voice activated powered wheelchair •WHEN user speaks commands, a microphone in a throat detects the vibration of vocal cord •UNSTABLE for powered wheelchair control in noisy
- 8. Head control •WHEELCHAIR controlled by head control device •NOT good for the users with bad head movement ability
- 9.
- 10. TONGUE and mouth occupies the major part MUSCLE fibers in tongue is similar to heart muscle fibers LOW rate of Perceived Exertion DIRECTLY connected to the brain HIDDEN inside mouth will give a certain degree of privacy
- 11.
- 12. TDS –Overview… SMALL permanent magnet pierced on the tongue ARRAY of Hall-effect magnetic sensors senses the magnetic field SENSORS mounted on a dental retainer and attached on the outside of the teeth SMALL batteries are intended to power
- 13. POWER management circuitry scans through the sensors and turn on one at a time TDM analog output are digitized, modulated and transmitted to the external control unit through a wireless link FROM their signals are demodulated, demultiplexed to extract individual sensor output BY processing these output command is identified
- 14.
- 15. BUILD on a face shield FUNCTION is to directly emulate mouse pointing and selection function with the tongue movement SIX COMMANDS : up, down, right, left, single click & double click SSP algorithm running in background
- 16. MAGNETIC TRACER : small, cylindrical, rare-earth permanent magnet PAIR of two-axis magnetic field sensor modules- each contained a pair of orthogonal magneto-inductive sensor THREE AXIS MODULE: used as a reference electronic compass to minimize the effect of external magnetic field interference
- 17. CONTROL unit & reference compass hidden under face shield cap ENTIRE system was powered by a 3.3V coin-sized battery(CR2023)
- 19. SENSOR output where send serially to the ultralow- power MSP430 microcontroller MICROCONTROLLER took 11 samples from each sensor SAMPLES are arranged in a data frame and wirelessly transmitted to a PC across a 2.4 GHz wireless link established between nRF2401 transceivers
- 20. Sensor Signal Processingalgorithm DEVELOPED in MATLAB environments 2 phases: Training & Testing •TRAINING : USES principal components analysis (PCA) to extract the most important features of the sensor output waveforms for each specific command • USER repeats each of the six designated commands 10 times in 3-second intervals
- 21. • TOTAL of 12 samples (3 per sensor) are recorded in 12-variable vectors • THE PCA-based feature-extraction algorithm calculates the eigenvectors and Eigen values based on the 12-variable vectors • THREE eigenvectors with the largest Eigen values are then chosen to set up the feature matrix [v1, v2, v3]
- 22. • BY multiplying the training vectors with the feature matrix, the SSP algorithm forms a cluster (class) of 10 data points from training for each specific command •TESTING: k-nearest neighbour (kNN) classifier is then used in real time to evaluate the proximity of the incoming data points to the clusters formed earlier in the training phase
- 23. • kNN starts at the incoming new data point and inflates an imaginary sphere around that data point until it contains a certain number (k) of the nearest training data points • IT associates the new data point to the command that has the majority of the training data points inside that spherical region.
- 24. AFTER finding theintended user command, the mouse pointer starts moving slowly in the selected direction FOR faster access the user can hold his or her tongue in the position of the issued command and the pointer will gradually accelerate until it reaches a certain maximum velocity
- 25. Training Session GRAPHICAL user interface (GUI) prompted the user to define each command by moving his tongue from its resting position to the corresponding command position when the command light was on and returning it back to the resting position when the light went off
- 26. THIS procedure was repeated 10 times for the entire set of six commands plus the tongue resting position, resulting in a total of 70 training data points.
- 27. Implementation - Controlling a Powered Wheel Chair
- 30. Block diagram &hardware component SYSTEM consists of the hall effect sensors (A1231), MSP430 microcontroller, H-Bridge driver (SN754410)
- 31. 4 ADC channels of MSP430 convert analog signals from sensors to digital for processing BASED on the processed information microcontroller drives the driver IC DRIVER IC drives the DC powered wheelchair
- 32. SENSOR- Linear Bipolar Hall effect sensor TRANSDUCER which varies its output voltage in response to changes in magnetic field •VOLTAGE:4.5V-5.5V •OUTPUT is an analog voltage that vary from 0-5V •THAT is converted to a digital value by ADC
- 33. H-BRIDGE DRIVER : used to drive the wheelchair SN7454410:BidirectionalH-Bridge driver 5V power supply is given SPEED of the wheelchair is controlled by the PWM signals from MSP430
- 34. •4 SWITCHES •WHEN Aand D are closed, a positive voltage applied across the motor •WHEN B and C are closed , voltage is reversed, allowing reverse operation of the motor
- 35. DC Motors and Gear Box •5V DC motors with a gearbox for driving the vehicle •USED 2 motors: Left and right •LEFT motor at rest and right in motion turn vehicle left and vice versa •FOR forward and backward motion both the motors are driven in same direction with same speed
- 36. Software Components ADC10(Analog to Digital converter) •MSP430G2252 has 8 ADC channels of 10 bit each •ADC10 convert the analog signal from the sensor into digital value •WE get the value ranging from 0-1023
- 38. Pulse Width Modulation •MODULATING technique which generates variable width pulses is used to vary the speed of the motor
- 39. Algorithm •ROUND ROBIN algorithm that implements polling technique •SENSOR values are always read one after other in a continuous loop •FEW threshold values set for each sensor •IF reading from sensors reaches the threshold value program triggers the change of state
- 41. S1 & S2 front sensor , S3 & S4 back sensors WHITE circle in the middle- resting position of tongue DIFFERENT flag values for each sensor IF sensor is high, then the flag value remains high until the operation is performed, then it is reset again
- 42. f1 - S1 > 750 // forward f2 - S2 > 600 // backward f3 - S3 > 700 // speed increase f4 - S4 > 700 // speed decrease f5 - S1 > 600 & S3 > 600 // turn left f6 - S2 > 600 & S4 > 600 // turn right
- 43. Advantages SIMPLE to implement , low cost, easy to operate flexible NO surgery needed OFFERS better privacy to the user
- 44. Drawbacks USERS should avoid inserting ferromagnetic objects in their mouth MAGNETIC tracer should be removed if the user is undergoing MRI
- 46.