Project Description:

The schematic above shows the general layout of the sensors used in our system. Laminated between two 0.02 inch sheets of clear Type 1 PVC (polyvinyl chloride) heavy-duty film sheets are the insole sensors: the insole sensors: four FSR pads (force sensitive resistors that measure continuous pressure), two PVDF strips (polyvinylidine fluoride, a piezoelectric material that measures dynamic pressure), and two pairs of resistive bend sensors. The FSRs provide a coarse measurement of the pressure distribution beneath the foot, and are placed with two beneath the heel (medially and laterally) and two located forward, behind the toes (one beneath the first metatarsal head, and another beneath the fourth and fifth metatarsal head). The PVDF strips provide dynamic information about "heel strike" and "toe off," and are located beneath the heel and the beneath the hallux (big toe). Each pair of bend sensors provides information about bi-directional bend (each pair consists of two bend sensors which are placed back to back). One pair provides information about the extent of plantarflexion or dorsiflexion at the ankle, hence is located at the back of the shoe and inserted into an ankle strap so that it bends around the rear edge of the shoe as the ankle tilts. The other pair of bend sensors provides information about the extent of plantarflexion or dorsiflexion at the metatarsals during walking, hence is located at the forward portion of the insole.

The shoe attachment contains three stacked circuit boards, the power supply (currently a 9V battery), and the antenna for wireless transmission. The mass of the current prototype is approximately 200 g. The circuit boards are modular components designed to be stacked atop each other (mating via a pair of commonly-placed connectors), and each address specific needs: one contains the conditioning electronics for the insole sensors, a second has the microcontroller that collects data from the other boards together with the wireless transceiver, and a third is a compact inertial measurement unit (IMU) that has a full set of three gyroscopes and three accelerometers to measure angular velocity and linear acceleration about three axes. Any redundancy in these measurements will be used to improve detection and reduce errors. A fourth board now in final development will provide capacitive sensing to measure the height of the shoe sole above ground, a sonar to measure the distance to the ground from the heel, and another sonar for measuring both distance and angle between shoes.

A simple TDMA scheme allows both shoes to share a single RF channel; the embedded transceivers on the SIGSS processor card are capable of operating at 115 kbps, allowing all 18 parameters from both shoes to be updated at better than 60 Hz.


Images:

The assembly on a shoe.   ||   Shoe attachment.   ||   Insoles.   ||   Circuit boards.   ||   Basestation.
Apologies, these are old photos! New ones should appear ~end of Feb '04

Papers:

Gait Analysis Using a Shoe-Integrated Wireless Sensor System, Bamberg, S.J.M., Benbasat A.Y., Scarborough D.M., Krebs D.E., Paradiso J.A., IEEE Transactions on Information Technology in Biomedicine, Vol. 12, No. 4, July 2008, pp. 413-423. (Best Paper Award)
This is a comprehensive paper describing the system and the results of our biomotion analysis.

Paradiso, J.A., Morris, S.J., Benbasat, A.Y., Asmussen, E., "Interactive Therapy with Instrumented Footwear," in the Proc. of the ACM Conference on Human Factors and Computing Systems (CHI 2004), Extended Abstracts, Vienna, Austria, April 27-29, 2004, pp. 1341-1343.
A short CHI paper on using the gait shoe to produce musical feedback for physical therapy and rehab.

Stacy J. Morris, Joseph A. Paradiso. "Shoe-integrated sensor system for wireless gait analysis and real-time feedback." Proceedings of the 2nd Joint IEEE EMBS (Engineering in Medicine and Biology Society) and BMES (the Biomedical Engineering Society) Conference, October, 2002, pp. 2468-2469.
This short paper describes the overall design concept, and was presented at the Second Joint IEEE/EMBS and BMES Joint Meeting. [EMBS=Engineering in Medicine and Biology Society; BMES=BioMedical Engineering Society]

S.J. Morris and J.A. Paradiso, "A Compact Wearable Sensor Package for Clinical Gait Monitoring." Offspring Vol. 1, No. 1, pp. 7-15, January 31, 2003.
A paper we wrote for the Motorola Offspring Journal, describing the overall sensor package.

Benbasat, A.Y., Morris, S.J, and Paradiso, J.A. "A Wireless Modular Sensor Architecture and its Application in On-Shoe Gait Analysis." In the Proceedings of the 2003 IEEE International Conference on Sensors, October 21-24, Toronto, Ontario, pp. 1086-1091.
A paper presented at IEEE Sensors 2003. This one describes the hardware in detail, and discusses some early results.

Stacy J Morris "A Shoe-Integrated Sensor System for Wireless Gait Analysis and Real-Time Therapeutic Feedback," PhD Thesis, MIT Mechanical Engineering Dept., June 2004,
Stacy's PhD Thesis on the Gait Shoe.

Video Clips:

General:
A quicktime video clip showing a subject walking with one shoe on the right foot and the corresponding stripchart outputs. [This video requires Quicktime 6 (free).]

A quicktime video clip showing a subject during testing in the MGH Biomotion Lab. The subject is wearing the Biomotion Lab equipment on eleven body segments and our shoe on the right foot.

Rhythmic Auditory Stimulation:
This work was done with in conjunction with Erik Asmussen, an undergraduate from Colgate, who worked the implementation of the auditory feedback during the summer of 2003 and during January 2004.

Clips from Erik's work in January 2004:

Clips from Erik's work in Summer 2003:

Links:

The Homepage for the Responsive Environments Group
The Projects Page for the Responsive Environments Group
Stacy Morris' current homepage - she is currently on faculty at the University of Utah