Nuit Blanche: darpa

Showing posts with label darpa. Show all posts

Thursday, October 02, 2008

CS: Manifold Models for Signals and Images, Thresholded Basis Pursuit, Maximum Entropy Method in CS Reconstruction and DARPA mathematical challenges.

We have three papers today related to different reconstruction techniques.

The first one refers back to the manifold signal processing we have seen before being developed by Michael Wakin in his thesis. While there is an issue of non-differentiability in images featuring occluding elements, the smooth manifold approach is somehow ideally fitted with that of textures. The paper by Gabriel Peyre entitled Manifold Models for Signals and Images looks at this problem. The abstract reads:

This article proposes a new class of models for natural signals and images. The set of patches extracted from the data to analyze is constrained to be close to a low dimensional manifold. This manifold structure is detailed for various ensembles suitable for natural signals, images and textures modeling. These manifolds provide a low-dimensional parametrization of the local geometry of these datasets. These manifold models can be used to regularize inverse problems in signal and image processing. The restored signal is represented as a smooth curve or surface traced on the manifold that matches the forward measurements. A manifold pursuit algorithm computes iteratively a solution of the manifold regularization problem. Numerical simulations on inpainting and compressive sensing inversion show that manifolds models bring an improvement for the recovery of data with geometrical features.

I note the development of a manifold based greedy algorithm called manifold pursuit.

The second paper is Thresholded Basis Pursuit: Quantizing Linear Programming Solutions for Optimal Support Recovery and Approximation in Compressed Sensing by Venkatesh Saligrama, Manqi Zhao. The abstract reads:

We consider the classical Compressed Sensing problem. We have a large under-determined set of noisy measurements Y=GX+N, where X is a sparse signal and G is drawn from a random ensemble. In this paper we focus on a quantized linear programming solution for support recovery. Our solution of the problem amounts to solving $\min \|Z\|_1 ~ s.t. ~ Y=G Z$, and quantizing/thresholding the resulting solution $Z$. We show that this scheme is guaranteed to perfectly reconstruct a discrete signal or control the element-wise reconstruction error for a continuous signal for specific values of sparsity. We show that in the linear regime when the sparsity, $k$, increases linearly with signal dimension, $n$, the sign pattern of $X$ can be recovered with $SNR=O(\log n)$ and $m= O(k)$ measurements. Our proof technique is based on perturbation of the noiseless $\ell_1$ problem. Consequently, the achievable sparsity level in the noisy problem is comparable to that of the noiseless problem. Our result offers a sharp characterization in that neither the $SNR$ nor the sparsity ratio can be significantly improved. In contrast previous results based on LASSO and MAX-Correlation techniques assume significantly larger $SNR$ or sub-linear sparsity. We also show that our final result can be obtained from Dvoretsky theorem rather than the restricted isometry property (RIP). The advantage of this line of reasoning is that Dvoretsky's theorem continues to hold for non-singular transformations while RIP property may not be satisfied for the latter case.

I note the independence of their approach from the Restricted Isometry Property.

Finally, the third paper is about using a maximum entropy method in A New Reconstruction Approach to Compressed Sensing by Tianjing Wang and Zhen Yang. The abstract reads:

Compressed sensing is a new concept in signal processing where one seeks to minimize the number of measurements to be taken from signals while still retaining the information necessary to approximate them well. Nonlinear algorithms, such as norm optimization problem, are used to reconstruct the signal from the measured data. This paper proposes a maximum entropy function method which intimately relates to homotopy method as a computational approach to solve the optimization problem. Maximum entropy function method makes it possible to design random measurements which contain the information necessary to reconstruct signal with accuracy. Both the theoretical evidences and the extensive experiments show that it is an effective technique for signal reconstruction. This approach offers several advantages over other methods, including scalability and robustness.

Finally, DARPA has put out a research request for proposals entitled Mathematical Challenges. Compressive Sensing has some obvious bearing on some of these challenges, to name a few I'd go for 6, 7, 8, 10 and 15.

Mathematical Challenge One: The Mathematics of the Brain

Develop a mathematical theory to build a functional model of the brain that is mathematically consistent and predictive rather than merely biologically inspired.

Mathematical Challenge Two: The Dynamics of Networks

Develop the high-dimensional mathematics needed to accurately model and predict behavior in large-scale distributed networks that evolve over time occurring in communication, biology and the social sciences.

Mathematical Challenge Three: Capture and Harness Stochasticity in Nature

Address Mumford’s call for new mathematics for the 21^st century. Develop methods that capture persistence in stochastic environments.

Mathematical Challenge Four: 21st Century Fluids

Classical fluid dynamics and the Navier-Stokes Equation were extraordinarily successful in obtaining quantitative understanding of shock waves, turbulence and solitons, but new methods are needed to tackle complex fluids such as foams, suspensions, gels and liquid crystals.

Mathematical Challenge Five: Biological Quantum Field Theory

Quantum and statistical methods have had great success modeling virus evolution. Can such techniques be used to model more complex systems such as bacteria? Can these techniques be used to control pathogen evolution?

Mathematical Challenge Six: Computational Duality

Duality in mathematics has been a profound tool for theoretical understanding. Can it be extended to develop principled computational techniques where duality and geometry are the basis for novel algorithms?

Mathematical Challenge Seven: Occam’s Razor in Many Dimensions

As data collection increases can we “do more with less” by finding lower bounds for sensing complexity in systems? This is related to questions about entropy maximization algorithms.

Mathematical Challenge Eight: Beyond Convex Optimization

Can linear algebra be replaced by algebraic geometry in a systematic way?

Mathematical Challenge Nine: What are the Physical Consequences of Perelman’s Proof of Thurston’s Geometrization Theorem?

Can profound theoretical advances in understanding three dimensions be applied to construct and manipulate structures across scales to fabricate novel materials?

Mathematical Challenge Ten: Algorithmic Origami and Biology

Build a stronger mathematical theory for isometric and rigid embedding that can give insight into protein folding.

Mathematical Challenge Eleven: Optimal Nanostructures

Develop new mathematics for constructing optimal globally symmetric structures by following simple local rules via the process of nanoscale self-assembly.

Mathematical Challenge Twelve: The Mathematics of Quantum Computing, Algorithms, and Entanglement

In the last century we learned how quantum phenomena shape our world. In the coming century we need to develop the mathematics required to control the quantum world.

Mathematical Challenge Thirteen: Creating a Game Theory that Scales

What new scalable mathematics is needed to replace the traditional Partial Differential Equations (PDE) approach to differential games?

Mathematical Challenge Fourteen: An Information Theory for Virus Evolution

Can Shannon’s theory shed light on this fundamental area of biology?

Mathematical Challenge Fifteen: The Geometry of Genome Space

What notion of distance is needed to incorporate biological utility?

Mathematical Challenge Sixteen: What are the Symmetries and Action Principles for Biology?

Extend our understanding of symmetries and action principles in biology along the lines of classical thermodynamics, to include important biological concepts such as robustness, modularity, evolvability and variability.

Mathematical Challenge Seventeen: Geometric Langlands and Quantum Physics

How does the Langlands program, which originated in number theory and representation theory, explain the fundamental symmetries of physics? And vice versa?

Mathematical Challenge Eighteen: Arithmetic Langlands, Topology, and Geometry

What is the role of homotopy theory in the classical, geometric, and quantum Langlands programs?

Mathematical Challenge Nineteen: Settle the Riemann Hypothesis

The Holy Grail of number theory.

Mathematical Challenge Twenty: Computation at Scale

How can we develop asymptotics for a world with massively many degrees of freedom?

Mathematical Challenge Twenty-one: Settle the Hodge Conjecture

This conjecture in algebraic geometry is a metaphor for transforming transcendental computations into algebraic ones.

Mathematical Challenge Twenty-two: Settle the Smooth Poincare Conjecture in Dimension 4

What are the implications for space-time and cosmology? And might the answer unlock the secret of “dark energy”?

Mathematical Challenge Twenty-three: What are the Fundamental Laws of Biology?

This question will remain front and center for the next 100 years. DARPA places this challenge last as finding these laws will undoubtedly require the mathematics developed in answering several of the questions listed above.

Monday, January 28, 2008

Compressed Sensing: Hardware Implementations in Computational Imaging, Coded Apertures and Random Materials

[Update Nov. 08: I have listed most of the Compressed Sensing Hardware in the following page]

I have mentioned some realization of hardware implementations of Compressed Sensing (or Compressive Sensing or Compressive Sampling) before ([L1], [L2],[L3],[L4],[L5],[L6]). However owing to my ignorance, I did not give full credit to some work that made the most contribution to the subject. I think it had mostly with the fact that some websites were not up at the time I looked at the issue or that too many articles required full access to some journals (some still do). Much work on the subject of Optical Compressed Sensing has been performed at the Duke Imaging and Spectroscopy Program or DISP led by David Brady and at the Optical Computing and Processing Laboratory led by Mark Neifeld at the University of Arizona.

To give some perspective on this new and evolving field here is the story is told by the Duke researchers in Compressive Imaging Sensors [1]

An optical image has been understood as an intensity field distribution representing a physical object or group of objects. The image is considered two dimensional because the detectors are typically planary, although the objects may not. This understanding of the optical intensity field as the image has persisted even as electronic focal planes have replaced photochemical films. Lately, however, more imaginative conceptions of the relationship between the detected field and the reconstructed image have emerged. Much of this work falls under the auspices of the “computational optical sensing and imaging”,1 which was pioneered in Cathey and Dowski’s use of deliberate image aberrations to extend the depth of field2, 3 and by computed spectral tomography as represented, for example, in the work by Descour and Derniak.4 More recently, both extended depth of field and spectral features in imaging systems have been considered by many research groups. Images of physical objects have many features, such as lines and curves as well as areas separated or segmented by lines and curves. The most fundamental feature of images is the fascinating fact that an image is not an array of independent random data values. Tremendous progress has been made in the past decade in feature extraction and compression of images via post digital processing. Only recently has intelligent sampling and compression at the physical layer become a major interest. The work of Neifeld is particularly pioneering in this regard.5, 6. The DISP group at Duke University has also focused in several studies on data representation at the optical sampling layer and on physical layer compression.7–12 The interest in data compression at physical layer is also encouraged by the mathematical results by Donoho et al., who measure general functionals of a compressible and discretized function and recover n values from O(n1/4 log5/2(n)) measurements. In particular, the 1-norm of the unknown signal in its representation with respect to an orthonormal basis is used as the minimization objective, subject to a condition on the sparsity in the representation coefficients.13, 14 Rapid progress along these lines by Candes, Baraniuk and others is summarized in publications on line www-dsp.rice.edu/CS/.

And therefore, it is becoming obvious that in the optics world, researchers have known for some time that one could get more information out of scenes as long as a physical layer allowed some type of interference between the signal and some physical device (preferably random). Before this entry, I had mentioned the Hyperspectral Imager at Duke and the page of Ashwin Wagadarikar but I had not seen the full list of publications at DISP that lists most of their work for the past five years on the subject. In line with the Random Lens Imager at MIT and the ability to detect movement as mentioned by Rich Baraniuk, here some articles that caught my eyes:

Multiple order coded aperture spectrometer by S. D. Feller, Haojun Chen, D. J. Brady, Michael Gehm, Chaoray Hsieh, Omid Momtahan, and Ali Adibi [2]. The abstract reads:

We introduce a multiple order coded aperture (MOCA) spectrometer. The MOCA is a system that uses a multiplex hologram and a coded aperture to increase the spectral range and throughput of the system over conventional spectrometers while maintaining spectral resolution. This results in an order of magnitude reduction in system volume with no loss in resolution.

This is, I believe, a follow-up of the work on Coded Aperture Snapshot Spectral Imaging (CASSI) mentioned before. At the end of this page there is a comparison between the two types of CASSI concept tried by the DISP group. Very informative. Then there is this series of papers in reference structure imaging, i.e. put something well designed in between the imager and the object and try to see what properties can be detected. In this case, they look at tracking objects:

Reference structure tomography, by David J. Brady, Nikos P. Pitsianis, and Xiaobai Sun, [3] The abstract reads:

Reference structure tomography (RST) uses multidimensional modulations to encode mappings between radiating objects and measurements. RST may be used to image source-density distributions, estimate source parameters, or classify sources. The RST paradigm permits scan-free multidimensional imaging, data-efficient and computation-efficient source analysis, and direct abstraction of physical features. We introduce the basic concepts of RST and illustrate the use of RST for multidimensional imaging based on a geometric radiation model.

Lensless sensor system using a reference structure by P. Potuluri, U. Gopinathan, J. R. Adleman, and D. J. Brady. The abstract reads:

We describe a reference structure based sensor system for tracking the motion of an object. The reference structure is designed to implement a Hadamard transformation over a range of angular perspectives. We implemented a reference structure with an angular resolution of 5o and a field of view of 40o

But then, after putting a well known object between the object and the imager, they insert random objects in Imaging with random 3D reference structures, by P. Potuluri, M. Xu, and D. J. Brady. The abstract reads:

Three dimensional (3D) reference structures segment source spaces based on whether particular source locations are visible or invisible to the sensor. A lensless 3D reference structure based imaging system measures projections of this source space on a sensor array. We derive and experimentally verify a model to predict the statistics of the measured projections for a simple 2D object. We show that the statistics of the measurement can yield an accurate estimate of the size of the object without ever forming a physical image. Further, we conjecture that the measured statistics can be used to determine the shape of 3D objects and present preliminary experimental measurements for 3D shape recognition.

and in Imaging with random 3D reference structures by Prasant Potuluri, Mingbo Xu and David J. Brady

The abstract reads:

We describe a sensor system based on 3D ‘reference structures’ which implements a mapping from a 3D source volume on to a 2D sensor plane. The reference structure used here is a random three dimensional distribution of polystyrene beads.We show how this bead structure spatially
segments the source volume and present some simple experimental results of 2D and 3D imaging.

And so they begin to detect size, shape and motion of objects! The shape feature has also been looked at by some folks at Rice in a small course on Compressed Sensing (course given on CNX.org).

The motion is studied in Coded apertures for efficient pyroelectric motion tracking, by U. Gopinathan, D. J. Brady, and N. P. Pitsianis

The abstract reads

Coded apertures may be designed to modulate the visibility between source and measurement spaces such that the position of a source among N resolution cells may be discriminated using logarithm of N measurements. We use coded apertures as reference structures in a pyroelectric motion tracking system. This sensor system is capable of detecting source motion in one of the 15 cells uniformly distributed over a 1.6microns × 1.6microns domain using 4 pyroelectric detectors.

The size and shape are investigated in Size and shape recognition using measurement statistics and random 3D reference structures by Arnab Sinha and David J. Brady

The abstract reads

Three dimensional (3D) reference structures segment source spaces based on whether particular source locations are visible or invisible to the sensor. A lensless 3D reference structure based imaging system measures projections of this source space on a sensor array. We derive and experimentally verify a model to predict the statistics of the measured projections for a simple 2D object. We show that the statistics of the measurement can yield an accurate estimate of the size of the object without ever forming a physical image. Further, we conjecture that the measured statistics can be used to determine the shape of 3D objects and present preliminary experimental measurements for 3D shape recognition.

A similar argument was used by the folks at University of Arizona to obtain Superresolution when they looked at the pseudorandom phase-enhanced lens (PRPEL) imager in Pseudorandom phase masks for superresolution imaging from subpixel shifting by Amit Ashok and Mark A. Neifeld [9]. The abstract reads:

We present a method for overcoming the pixel-limited resolution of digital imagers. Our method combines optical point-spread function engineering with subpixel image shifting. We place an optimized pseudorandom phase mask in the aperture stop of a conventional imager and demonstrate the improved performance that can be achieved by combining multiple subpixel shifted images. Simulation results show that the pseudorandom phase-enhanced lens (PRPEL) imager achieves as much as 50% resolution improvement over a conventional multiframe imager. The PRPEL imager also enhances reconstruction root-mean-squared error by as much as 20%. We present experimental results that validate the predicted PRPEL imager performance.

The idea is to, through the use of a random phase materials, spread out the Point Spread Function (it is generally a point/dirac in normal cameras) so that it is bigger and more easily delineated from other points. The expectation is that the diffraction limit is pushed since now one can delineate more easily one "spread" peak from another one (the two peaks are not spread symmetrically thanks to the random materials). The ability to have higher resolution will then come from the ability in compressed sensing to find the sparsest dictionary of Point Spread Functions that can explain the image.

Some more explanative figures (see above) can be found in Imager Design using Object-Space Prior Knowledge a presentation at IMA 2005 by Mark Neifeld.

All in all, it seems to me that the major issue when designing these random imagers is the calibration issue that seems to be very cumbersome. Is there a way to do this faster ? Can Machine learning help ?

On a related note, Dharmpal Takhar will defend his thesis on Compressed Sensing for Imaging Applications. It'll be on Monday, February 4, 2008, from 10:00 AM to 12:00 PM in 3076 Duncan Hall at Rice University.

His abstract is:

Compressed sensing is a new sampling theory which allows reconstructing signals using sub-Nyquist measurements/sampling. This can significantly reduce the computation required for image/video acquisition/encoding, at least at the sensor end. Compressed sensing works on the concept of sparsity of the signal in some known domain, which is incoherent with the measurement domain. We exploit this technique to build a single pixel camera based on an optical modulator and a single photosensor. Random projections of the signal (image) are taken by optical modulator, which has random matrix displayed on it corresponding to the measurement domain (random noise). This random projected signal is collected on the photosensor and later used for reconstructing the signal. In this scheme we are making a tradeoff between the spatial extent of sampling array and a sequential sampling over time with a single detector. In addition to this method, we will also demonstrate a new design which overcomes this shortcoming by parallel collection of many random projections simultaneously. Applications of this technique in hyperspectral and infrared imaging will be discussed.

This is going to be interesting, I can't wait to see how the random projections are gathered simultaneously. Good luck Dharmpal!

References:
[1] N. Pitsianis, D. Brady, A. Portnoy, X. Sun, M. Fiddy, M. Feldman, R. TeKolste, Compressive Imaging Sensors, ," Proceedings of SPIE. Vol. SPIE-6232,pp. 43-51. (2006)

[2] Multiple order coded aperture spectrometer, S. D. Feller, Haojun Chen, D. J. Brady, M. E. Gehm, Chaoray Hsieh, Omid Momtahan, and Ali Adibi , Optics Express, Vol. 15, Issue 9, pp. 5625-5630

[3] David J. Brady, Nikos P. Pitsianis, and Xiaobai Sun, Reference structure tomography, J. Opt. Soc. Am. A/Vol. 21, No. 7/July 2004

[4] P. Potuluri, U. Gopinathan, J. R. Adleman, and D. J. Brady, ‘‘Lensless sensor system using a reference structure,’’ Opt. Express 11, 965–974 (2003).

[5] Coded apertures for efficient pyroelectric motion tracking, U. Gopinathan, D. J. Brady, and N. P. Pitsianis Opt. Express 11, 2142–2152 (2003).

[6] Size and shape recognition using measurement statistics and random 3D reference structures, Arnab Sinha and David J. Brady

[8] Pseudorandom phase masks for superresolution imaging from subpixel shifting by Amit Ashok and Mark A. Neifeld Applied Optics, Vol. 46, Issue 12, pp. 2256-2268

[9] A. Ashok and M. A. Neifeld, " Pseudorandom phase masks for superresolution imaging from subpixel shifting," Appl. Opt. 46, 2256-2268 (2007)

Saturday, November 03, 2007

DARPA Urban Challenge: It's today

We did not qualify for the semi-finals but others did. DARPA has finally selected the teams that will compete in the Urban Challenge Final Event on Saturday, November 3 at the former George Air Force Base in Victorville, Calif. The teams will attempt to complete a complex 60-mile urban course with live traffic in less than six hours. The finalists will operate on the course roads with approximately 50 human-driven traffic vehicles. Speed is not the only factor in determining the winners, as vehicles must also meet the same standards required to pass the California DMV road test. The Urban Challenge Final Event is to take place today (November 3) and will be webcast live at http://www.grandchallenge.org starting at 7:30 am PT. Grounds will open at 6:00 AM PT for spectators, but the opening ceremony will begin at 7:30 AM, and vehicles will begin to launch at 8:00 AM.

Watch the Race here, Status Board is here, Live Video is here, Tracking Map is here.

A brief highlight video of National Qualifying Event (semi-final) operations is available for download here: High (27MB) Med (7.5MB) Low (3MB)

Vehicles are being tested in three test areas to evaluate their ability to operate with live traffic, make safe left turns across moving traffic, and pull out at T-intersections with cars arriving from both directions. Vehicles also have to follow narrow winding roads, avoiding parked cars and other obstructions, maneuver into a designated parking spot and negotiate 4-way intersections and road-blocks.

List of teams that did qualify for the Final Event can be found here.
List of teams that did not qualify for the Final Event can be found here.

I personally will not watch, I am still sore from this.

Saturday, April 14, 2007

It's not about the sensor, it's about the target

Terry Tao just wrote an entry in his blog trying to explain compressed sensing using the lens of the one-pixel camera from Rice. It's been my experience that the compressed sensing results are so unusual in terms of engineering, that, as proved by the length of Terry's post, there is a lot of explaining to do. On top of it, the use of the wording "compressed sensing" or "compressive sampling" or "compressive sensing" automatically gets engineers to think of compression as implemented in the JPEG/JPEG2000 sense (transform coding), so much of the explanation on compressed sensing goes toward explaining why this is strictly not that (transform coding). So while most people's question is whether this new type of camera change the digital world, it does not come accross that the one of the force of the technique is the tremendous simplification and robustness of detection algorithms. This is not trivial as image processing always involved a lot of parameter tweaking in order to obtain "the best result". It should help.

Sunday, April 08, 2007

DARPA Urban Challenge: Unveiling our algorithm

In a previous entry, I mentioned the fact that we are unveiling our strategy.unveiling our strategy for our entry in the DARPA Urban Challenge (DARPA Urban Challenge is about driving at a record pace in some urban environment past many urban difficulties, including no GPS capability). This was not accurate, in fact, we really are going to unveil our algorithm. You'll be privy of the development quirks and everything that goes on implementing an algorithm that has to respond on-line to a challenging situation. I'll be talking on the history of why we are choosing specific algorithms over others. I will specifically talk more about manifold-based models for decision making in the race and the use of techniques devised to produce a storage device of previous actions in order to produce some sort of supervised learning capability. In the previous race, we were eliminated early mostly because we were plagued with mechanical problems that most of us had never faced before (none of us had robotics background), we hope to go farther this time as the vehicle is OK. For reference, we have already shown some of our drive by wire program before as well. We made some of our data available before and I expect, as time permit to do the same as we go along. Because our entry is truely innovative, we are trying to balance not getting eliminated by passing every steps of the application and those innovation in the algorithm. However, since all of us are not interested in just an autonomous car, our emphasis will always be on the side of doing something that most other entries are not attempting such as using compressed sensing and robust mathematical techniques for instance.

Saturday, March 24, 2007

Driving on a manifold: unveiling our strategy

Our entry in DARPA Urban Challenge will feature compressed sensing as a way to reduce the dimensionality of our vision sensor. We will then have to infer the connection between our GPS track (RNDF and MDF) and the reduced parameters obtained from random projections.

Wednesday, March 14, 2007

Implementing Compressed Sensing in Applied Projects

We are contemplating using Compressed Sensing in three different projects:

The Hyper-GeoCam project: This is a payload that will be flown on the HASP platform in September. Last year, we flew a simple camera that eventually produced a 105 km panorama of New Mexico. We reapplied for the same program and have been given the OK for two payloads. The same GeoCam will be re-flown so that we can produce a breath taking panorama from 36 km altitude. The second payload is essentially supposed to be a hyperspectral imager on the cheap: i.e. a camera and some diffraction gratings allowing a fine decomposition of the reflected sun light from the ground. The project is called Hyper-GeoCam and I expect to implement a random lens imager such as the one produced at MIT. Tests will be performed on the SOLAR platform.
The DARPA Urban Challenge: We have a car selected in the track B: We do not have Lidars and need to find ways to navigate in an urban settings with little GPS availability. The autonomous car is supposed to be navigating in a mock town and follow the rules of the California traffic laws, that includes passing other cars.
Solving the Linear Boltzmann equation using compressed sensing techniques:The idea is that this equation has a known suite of eigenfunctions (called Case eigenfunctions) and because they are very difficult to use and expand from, it might be worth a try to look into the compressed sensing approach to see if it solves the problem more efficiently.

Wednesday, March 07, 2007

Simply Knowing Yourself

While reading this paper on the current illusion set forth by "better" classifiers, I could not help but remember Sebastian Thrun's presentation on winning DARPA first Grand Challenge.

While classifying is one thing, maybe quantifying one's state of knowledge or your state of ignorance might be more relevant.

Tuesday, March 06, 2007

It's the palm cooling, stupid.

When I was reading Tony Tether's interview on the cool glove, I could not shake the thought that it is connected to several areas of interest I have. In this Stanford paper, it is shown that cooling through the palms of your hand is really important for most physical exhaustive activity as well as for people who suffer from MS. The principle is that palms are the main radiators for the body.

What Heller and Grahn were seeing was the return trip: when externally applied heat shocked open the radiators in the cold palms of anesthesia patients, warmed blood was returned straight to the heart, and the body was reheated from the inside out. Applying a mild vacuum to the hand intensified this effect.

But this part of the entry stuck me

Grahn’s latest homemade version features soft vinyl against the hand instead of metal. One design challenge is obvious—how to create a vacuum-bearing glove flexible enough so that its wearers can use their hands, not just sit cooling their palms.

what he is describing is an element of an reversed advanced spacesuit.

This quote

Heller and Grahn have found in the lab that the temperature under which the radiators shut down in humans is highly individual.

strucks me as requiring some type of system to evaluate the radiator capacity for every potential customer. The RTX device using this concept is currently made by Avacore.

While reading this, I could not shake the fact that it was doing the reverse of the heat pipe glove and wonder how Bejan's work can be used to figure out an optimal cooling/heating solution that does not require a compressor.

Darpa's ability to innovate

Noah Shachtman interviews Tony Tether, the head of DARPA in this WIRED blog entry. Here is the quote I like:

People come to me from all over the world, and they look at our track record and what we're doing, and they want to know how they can make an organization like ours. I tell them it's simple. You just have to make sure the people don't stay there very long.

Tuesday, September 26, 2006

Cognitive Convergence

I am not blogging that much these days for several reasons. First, I am looking into how we can use some of the artificial intelligence techniques we are developing for the DARPA Urban Grand Challenge to diagnose and explore autism. I recently attended a talk by Hideki Kozima who uses of small robots like Keepon to evaluate the socialization of autistic kids. Take a look at the video here. After five to ten sessions with the Keepon, he could show the beginning of a joint attention development in autistic kids. This was very impressive.

I am also involved in the processing of the data we just received from our GeoCam on HASP. We have 4 GB of data to share with the rest of the world. We are developing a strategy on how to do this efficiently.

We are considering a run for the DARPA Urban Grand Challenge with Pegasus Bridge 1. Only this time, the mechanical aspect of the project will take a back seat to other technologies we are developing. We are interested in driver's gaze recording and supervised learning of road driving behavior. But more on this later...

Thursday, March 31, 2005

Test Site Visit

This is the site at the Southwest Research Institute where we will be demonstrating our vehicle.

Tuesday, March 15, 2005

DARPA Grand Challenge Team Progress

Thursday, March 10, 2005

Pegasus Bridge 1 enters DARPA's Grand Challenge 2005

Our entry in to DARPA's Grand Challenge

. This whole thing reminds me of a sentence from a famous speech but that no one remembers because it is too "local" and is often omitted in preference of the next sentence: Why does Rice play Texas ?

There is no strife, no prejudice, no national conflict in outer space as yet. Its hazards are hostile to us all. Its conquest deserves the best of all mankind, and its opportunity for peaceful cooperation may never come again. But why, some say, the moon? Why choose this as our goal? And they may well ask why climb the highest mountain. Why, 35 years ago, fly the Atlantic? Why does Rice play Texas? We choose to go to the moon. We choose to go to the moon in this decade and do the other things, not because they are easy, but because they are hard, because that goal will serve to organize and measure the best of our energies and skills, because that challenge is one that we are willing to accept, one we are unwilling to postpone, and one which we intend to win, and the others, too.

Wednesday, February 16, 2005

Cognitive Problems and Learning Robots

When I'll be talking at SCM next month, my talk will most likely make the connection between the type of studies currently undertaken in Robot learning and how they can be used towards diagnostics in people with cognitive problems. Generally, this knowledge is applied the other way around. An example of that approach can be found in this article on a Bayesian model of imitation in infants and robots. In it, one can read:

In particular, we intend to study the task of robotic gaze following. Gaze following is an important component of language acquisition: to learn words, a first step is to determine what the speaker is looking at, a problem solved by the human infant by about 1 year of age (Brooks & Meltzoff, 2002). We hope to endow robots with a similar capability.

This is fascinating because one of the reason kids with autism have difficulty learning is that their 'gazing' behavior is not optimal, which in effect stops them from learning by imitation. Another aspect of the connection between the two approaches is how imitation can be a quantifier of Autism at an early age. Video made by families of kids with autism show charateristic behaviors at an early age (9-12 months.) More recent papers show the behavior can be seen earlier at 4-6 months.

Tuesday, February 15, 2005

195 teams

The Darpa Grand Challenge has received 195 applications

Friday, February 04, 2005

We want to see everything and make sense of it

When writing proposals, it is always a good thing to look at past proposals and see if there is a trend. In particular, it is interesting to understand the slection mechanism by which the decision is made. Since it takes so much of a researcher's time to write a proposal, reading past debriefings is a good return on investment in answering the question: Will my idea go through the different tests of this organization. For instance, let say you want to contribute to a specific technology for the NRO through the Director's Innovative Initiative. Unlike some other agencies, this one provide breifings from previous years. It is a good thing. Unfortunately, it is all in a pdf format. In order to fully utilize this information, you really need to have some type of templates that uses all the remarks that were made in this briefing. If they are in pdf, you have to retype everything, which is a pain. Fortunately, Adobe allows a pdf-to-text converter to work over the web if the initial document is on the web. It is a nice little tool that should allow me to use the DII's debriefing present in my document so I can write a proposal accordingly. I believe there is a market for this type of application. An application that somehow reminds you of the essential points that need to be put in the document that you are producing. Large and small companies have boilerplates which have seen the test of time. But I think there is a market for small companies that just start entering new businesses.

Wednesday, January 26, 2005

Colloque "Mathematiques du Reel" / Colloquium on Real World Mathematics

Les inscriptions pour le Colloque "Mathematiques du Reel" sont ouvertes. Le Colloque est gratuit mais il faut s'inscrire. Cela se passe du lundi 21 au vendredi 25 mars 2005 dans les locaux de EdF R&D, 1 avenue du Général de Gaulle, 92141 Clamart, Salle Pierre Ailleret. Pour plus d'information sur le programme et les presentateurs peuvent se trouver surle site du colloque.

Registration for the Colloquium on Real World Mathematics is open. The colloquium is free but one has to regsiter. It will take place Monday through Friday March 21-25. The location is EdF R&D, 1 avenue du Général de Gaulle, 92141 Clamart in the Pierre Ailleret room. More information on the program and the speakers can be found here (the meeting will be held in French.)

Wednesday, October 13, 2004

Entering the DARPA Grand Challenge Race

After much time looking into it, we are ready to make this hobby a more serious hobby. I created a blog for our entry into the DARPA Grand Challenge race. Not much detail can be found there because with a small team, we'd rather concentrate on the most important thing, the vehicle and its autonomous driving system. Why Pegasus Bridge ? Because, on all accounts, we don't think we are building a car....