Electrical Project Proposals

Development of a 'Smart' Deep Brain Simulator

Overview

Deep brain stimulation (DBS) is used to treat Parkinson's Disease. Electrodes are surgically placed in the subthalamic nuclei of each patient, and continuous ~180 Hz electrical stimulation is delivered to this part of the brain. Symptoms of Parkinson's Disease are often dramatically improved (less tremor, stiffness and difficulty initiating movements). But no one knows how the therapy works. We believe that the stimulation releases neurotransmitters, glutamate within the subthalamic nucleus and dopamine at more distant sites in the motor control circuit. A self-sufficient brain machine interface that could sense changes in the neural environment (i.e. neurotransmitter levels) in its vicinity and accordingly direct electrical pulses of the correct amplitude and frequency through microelectrodes to a specific location in the subthalamic nucleus would create closed loop control of the stimulator and improve Parkinson's disease treatment and reduce DBS side effects. However, it is not clear which neurotransmitter is best to use as the feedback signal in a smart DBS system.

We believe that DBS causes glutamate release in the subthalamic nucleus and probably in the substantia nigra (where dopaminergic neurons reside), and that glutamate release, in turn, triggers the production of more dopamine, 'resetting' the motor control structures in the brain. To design DBS control algorithms, an understanding of glutamate and dopamine release as a function of DBS is necessary. We have already used a pseudo-random sequence of stimulation of the subthalamic nucleus (STN) in rats and the tools of system identification to determine a transfer function between the DBS input and the glutamate concentration (the output) and tested its accuracy and ability to predict experimental results. Now we are interested in determining the link between dopamine and glutamate.

Goals

1. Develop a LabVIEW-based experimental interface to control stimulation and assess neurotransmitter release.
2. Develop an interface for control of fast cycle voltammetry to measure dopamine levels moment by moment.
3. Develop adaptive mechanics for dynamic control of glutamate or dopamine release.
4. Use the tools of system identification to measure the transfer function between electrical stimulation and glutamate and dopamine in a rodent model of Parkinson's Disease.
5. Determine whether the transfer functions for glutamate and dopamine are correlated, and determine how this correlation is related to the loss of dopaminergic neurons in a rodent model of Parkinson's Disease.

Deliverables

1. An analysis of the limitations of the current technology available - what needs to happen in order to use smart DBS in humans?
2. A report comparing the benefits of smart DBS based on dopamine or glutamate - which neurotransmitter is the better target for any commercially viable 'smart' DBS system?
3. A fully functional algorithm-based experimental control model of a healthy motor control system written in LabVIEW.
4. A 'smart' DBS system prototype:
   1. The prototype will be based on glutamate or dopamine as demonstrated by targeted control of these neurotransmitter levels in the brains of individual rodents.
   2. The prototype will be adapted so that when it is used in a rodent with induced Parkinson's, it will maintain the level of the target neurotransmitter within a certain window.

Required Facilities

• Must be proficient with MATLAB programming
• Must have an interest in neuroscience
• LabVIEW experience helpful, but not necessary
• System identification knowledge helpful, but not necessary

Knowledge Areas Needed for Project

• Control systems
• Analog electronics
• Digital electronics
• Signal processing
• Software engineering
• Materials science
• Biomedical engineering
• Product design
• Statistical analysis
• Modeling and optimization

Proprietary Information and Confidentiality Requirements

• Sponsor accepts responsibility to discuss IP ownership directly with the student project team and project advisor

Formula Racing Hybrid IC Controls and Simulation

Overview

The engine of a hybrid vehicle differs from an ordinary vehicle in that it is not directly controlled by the driver. In a series hybrid drivetrain, the driver's input directly controls the electric motor, and the engine responds to the power demands of the drivetrain. This requires the implementation of an engine controller. The engine can be run in a variety of different modes. In order to determine the best mode of operation, the performance of the car must be analyzed in simulation so that the performance of each mode of operation can be determined. In addition, the engine requires a controller that is capable of maintaining proper engine operation under various load conditions.

This will be a very challenging project, given the complexity of the system and the relatively young age of the project. The simulation must be validated through testing, requiring a final product well in advance of the mid-March due date. It is expected that the system will allow the car to run in EV-only or IC-only mode. Furthermore, the car should be able to perform quite well in either mode.

Goals

This team will work closely with the energy storage team, generator team, and data acquisition team to develop a MATLAB simulation of the car. This code will be based on code developed last year. The results of these simulations will be used to compare energy accumulators as well as the optimum mode of operation for the engine. This team will then seek to develop an engine controller that is capable of maintaining proper engine operation under various loads. This controller may be constructed from scratch or from a modification to an existing off-the-shelf generator controller. This team will additionally be responsible for the design of any addition controller modules required to manage the power flow between blocks of the car.

Deliverables

By the first progress report, we expect the team to have an initial simulation of car performance. Based on these simulations, we expect this team to have chosen an optimum mode of operation for the engine which is well supported with decision matrices and simulation. By the end of the first term, we expect the team to have built a working throttle controller which is capable of maintaining proper engine operation under varying load conditions. The second term will be used for additional controller design, system redesign, data collection, and refinement. By the end of the second term, we expect a working control system for the vehicle which is not only able to maintain proper operation, but is able to shut down the vehicle under overstress or dangerous conditions. We expect full documentation on data collection, algorithm development, and controller implementation. Construction must be robust and professional; signals need to be provided to other systems in the car.

Required Facilities

• CAD - Cadence, SIMetrix, MATLAB
• Electrical - Standard lab facilities
• Mechanical - Solder station, assembly area

Knowledge Areas Needed for Project

• Machine design
• Control systems
• Analog electronics
• Digital electronics
• Signal processing
• Computer engineering
• Product design
• Engineering graphics (CAD/CAM)
• Project management

Proprietary Information and Confidentiality Requirements

None

The Super Intelligent Earplug

Overview

It goes without saying that classical music is dramatically different from all other musical styles. It may be less obvious to many people, though, that some of those differences are:

1. The music is almost always performed purely acoustically, without any amplification whatsoever (except when performing in outdoor venues) and
2. The music is often extremely flexible, and the musicians are continuously adjusting to each other (and the conductor) in order to stay together, and
3. The music vacillates frequently and rapidly between volume extremes and furthermore often has different instruments playing simultaneously both very loud and very soft.

In order for the classical music performer to be able to perform at their peak they need to be able to hear well both themselves and their colleagues at all times.

As the symphony orchestra has evolved over the last 200 years the number of instruments has increased dramatically, and the volume of many of those instruments has also increased. For many orchestral musicians it is now imperative that they wear ear protection if they expect to retain their hearing throughout their career. (The major symphony orchestras have experimented with sound baffles, meant to protect the more downstage musicians from the louder sounds of the more upstage players, with mixed results. There is, however, general agreement that they are unsightly to the audience.)

Unfortunately, any ear protection yet devised causes a significant loss of clarity of the ambient sounds, often making it difficult or impossible for the musicians to hear themselves. Additionally, the musicians are forced to insert and remove earplugs often during a performance, a distraction to themselves, their colleagues and the audience. The classical musician needs to be able to continue to hear clearly and well, yet have the total volume stay at manageable, non-damaging levels.

The Intelligent Earplug will enable the performing musician to hear well and clearly at all times by passing through to the ear-canal all safe sounds at a very high level of audio fidelity. For obvious reasons this needs to be accomplished without any noticeable latency. Those sounds that exceed the user-selectable threshold for loudness comfort will be reduced to the threshold level. The Intelligent Earplug will be as small, discrete and comfortable as conventional earplugs in order that its use will be transparent to the concert-going audience. It will operate with its own self-contained power source in order that it not require any wires, even if that means that it will need to be charged between uses.

Performers on some of the lower volume and lower pitched instruments (e.g. the cellos and basses) often find that they cannot monitor their own sound well at low volume, even when the overall volume of the orchestra is below the threshold for damage. In this situation one is often instinctively tempted to play slightly louder, resulting in a lack of blend within the section.

The Super-Intelligent Earplug will additionally allow the performer to mix into the ambient sound a small user-controllable amount of his/her own sound. The Super-Intelligent Earplug will add a wireless pickup, discretely mounted on the instrument, with the ability to transmit wirelessly to the Intelligent Earplug. In order that many musicians might simultaneously use this technology, it is important that the wireless transmitter and receiver be paired in a way that they can only communicate with each other.

Goals

Determine the best hardware and software configuration to maximize audio fidelity while minimizing power requirements for each of the three components.

Deliverables

Ideally, one working prototype with ideas for future improvements.

Knowledge Areas Needed for Project

• Analog electronics
• Digital electronics

Proprietary Information and Confidentiality Requirements

None