Skip to main content

Home/ Advanced Concepts Team/ Group items tagged trajectory optimization

Rss Feed Group items tagged

Luís F. Simões

Evolution of AI Interplanetary Trajectories Reaches Human-Competitive Levels - Slashdot - 4 views

  • "It's not the Turing test just yet, but in one more domain, AI is becoming increasingly competitive with humans. This time around, it's in interplanetary trajectory optimization. From the European Space Agency comes the news that researchers from its Advanced Concepts Team have recently won the Gold 'Humies' award for their use of Evolutionary Algorithms to design a spacecraft's trajectory for exploring the Galilean moons of Jupiter (Io, Europa, Ganymede and Callisto). The problem addressed in the awarded article (PDF) was put forward by NASA/JPL in the latest edition of the Global Trajectory Optimization Competition. The team from ESA was able to automatically evolve a solution that outperforms all the entries submitted to the competition by human experts from across the world. Interestingly, as noted in the presentation to the award's jury (PDF), the team conducted their work on top of open-source tools (PaGMO / PyGMO and PyKEP)."
  •  
    We made it to Slashdot's frontpage !!! :)
  •  
    Congratulations, gentlemen!
Ma Ru

An Inflationary Differential Evolution Algorithm for Space Trajectory Optimization - 7 views

  •  
    I was so shocked not to see Dario in the authors list!
  •  
    still practically as an ACT paper ... the first author was the first RF of the team and the one who suggested Dario ...
  •  
    Yeah I've figured it out from his CV at the end of the article after posting :)
eblazquez

[2106.09125] Convex Optimization for Trajectory Generation - 0 views

  •  
    Very cool recap document on state of the art sequential convex programming and lossless convexification techniques. It's written by the main authors who work on these, worth a look if you're into autonomous trajectory planning.
Dario Izzo

Optimal Control Probem in the CR3BP solved!!! - 7 views

  •  
    This guy solved a problem many people are trying to solve!!! The optimal control problem for the three body problem (restricted, circular) can be solved using continuation of the secondary gravity parameter and some clever adaptation of the boundary conditions!! His presentation was an eye opener ... making the work of many pretty useless now :)
  • ...13 more comments...
  •  
    Riemann hypothesis should be next... Which paper on the linked website is this exactly?
  •  
    hmmm, last year at the AIAA conference in Toronto I presented a continuation approach to design a DRO (three-body problem). Nothing new here unfortunately. I know the work of Caillau, although interesting what is presented was solved 10 years ago by others. The interest of his work is not in the applications (CR3BP), but in the research of particular regularity conditions that unfortunately make the problem limited practically. Look also at the work of Mingotti, Russel, Topputo and other for the (C)RTBP. Smart-One inspired a bunch of researchers :)
  •  
    Topputo and some of the others 'inspired' researchers you mention are actually here at the conference and they are all quite depressed :) Caillau really solves the problem: as a one single phase transfer, no tricks, no misconvergence, in general and using none of the usual cheats. What was produced so far by other were only local solutions valid for the particular case considered. In any case I will give him your paper, so that he knows he is working on already solved stuff :)
  •  
    Answer to Marek: the paper you may look at is: Discrete and differential homotopy in circular restricted three-body control
  •  
    Ah! with one single phase and a first order method then it is amazing (but it is still just the very particular CRTBP case). The trick is however the homotopy map he selected! Why this one? Any conjugate point? Did I misunderstood the title ? I solved in one phase with second order methods for the less restrictive problem RTBP or simply 3-body... but as a strict answer to your title the problem has been solved before. Nota: In "Russell, R. P., "Primer Vector Theory Applied to Global Low-Thrust Trade Studies," JGCD, Vol. 30, No. 2", he does solve the RTBP with a first order method in one phase.
  •  
    I think what is interesting is not what he solved, but how he solved the problem. But, are means more important than end ... I dunno
  •  
    I also loved his method, and it looked to me that is far more general than the CRTBP. As for the title of this post, OK maybe it is an exageration as it suggests that no solution was ever given before, on the other end, as Marek would say "come on guys!!!!!"
  •  
    The generality has to be checked. Don't you think his choice of mapping is too specific? he doesn't really demonstrate it works better than other. In addition, the minimum time choice make the problem very regular (i guess you've experienced that solving min time is much easier than mass max, optimality-wise). There is still a long way before maximum mass+RTBP, Topputo et al should be re-assured :p Did you give him my paper, he may find it interesting since I mention the homotopy on mu but for max mass:)
  •  
    Joris, that is the point I was excited abut, at the conference HE DID present solutions to the maximum mass problem!! One phase, from LEO to an orbit around the moon .. amazing :) You will find his presentation on line.... (according to the organizers) I gave him the reference to you paper anyway, but no pdf though as you did not upload it on our web pages and I could not find it in the web. So I gave him some bibliography I had with be from the russians, and from Russell, Petropoulos and Howell, As far as I know these are the only ones that can hope to compete with this guy!!
  •  
    for info only, my phd, in one phase: http://pdf.aiaa.org/preview/CDReadyMAST08_1856/PV2008_7363.pdf I prefered Mars than the dead rock Moon though!
  •  
    If you send me the pdf I can give it to the guy .. the link you gave contains only the first page ... (I have no access till monday to the AIAA thingy)
  •  
    this is why I like this Diigo thingy so much more than delicious ...
  •  
    What do you mean by this comment, Leopold? ;-) Jokes apart: I am following the Diigo thingy with Google Reader (rss). Obviously, I am getting the new postings. But if someone later on adds a comment to a post, then I can miss it, because the rss doesn't get updated. Not that it's a big problem, but do you guys have a better solution for this? How are you following these comments? (I know that if you have commented an entry, then you get the later updates in email.) (For example, in google reader I can see only the first 5 comments in this entry.)
  •  
    I like when there are discussions evolving around entries
  •  
    and on your problem with the RSS Tamas: its the same for me, you get the comments only for entries that you have posted or that you have commented on ...
LeopoldS

Characterizing Quantum Supremacy in Near-Term Devices - 2 views

shared by LeopoldS on 04 Sep 16 - No Cached
  •  
    google paper on quantum computers ... anybody with further insight on how realistic this is
  •  
    Not an answer to Leopold's question but here is a little primer on quantum computers for those that are (like me) still confused about what they actually do: http://www.dwavesys.com/tutorials/background-reading-series/quantum-computing-primer It give a good intuitive idea of the kinds of problems that an adiabatic quantum computer can tackle, an easy analogy of the computation and an explanation of how this get set up in the computer. Also, there is emphasis on how and why quantum computers lend themselves to machine learning (and maybe trajectory optimization??? - ;-) ).
Guido de Croon

Will robots be smarter than humans by 2029? - 2 views

  •  
    Nice discussion about the singularity. Made me think of drinking coffee with Luis... It raises some issues such as the necessity of embodiment, etc.
  • ...9 more comments...
  •  
    "Kurzweilians"... LOL. Still not sold on embodiment, btw.
  •  
    The biggest problem with embodiment is that, since the passive walkers (with which it all started), it hasn't delivered anything really interesting...
  •  
    The problem with embodiment is that it's done wrong. Embodiment needs to be treated like big data. More sensors, more data, more processing. Just putting a computer in a robot with a camera and microphone is not embodiment.
  •  
    I like how he attacks Moore's Law. It always looks a bit naive to me if people start to (ab)use it to make their point. No strong opinion about embodiment.
  •  
    @Paul: How would embodiment be done RIGHT?
  •  
    Embodiment has some obvious advantages. For example, in the vision domain many hard problems become easy when you have a body with which you can take actions (like looking at an object you don't immediately recognize from a different angle) - a point already made by researchers such as Aloimonos.and Ballard in the end 80s / beginning 90s. However, embodiment goes further than gathering information and "mental" recognition. In this respect, the evolutionary robotics work by for example Beer is interesting, where an agent discriminates between diamonds and circles by avoiding one and catching the other, without there being a clear "moment" in which the recognition takes place. "Recognition" is a behavioral property there, for which embodiment is obviously important. With embodiment the effort for recognizing an object behaviorally can be divided between the brain and the body, resulting in less computation for the brain. Also the article "Behavioural Categorisation: Behaviour makes up for bad vision" is interesting in this respect. In the field of embodied cognitive science, some say that recognition is constituted by the activation of sensorimotor correlations. I wonder to which extent this is true, and if it is valid for extremely simple creatures to more advanced ones, but it is an interesting idea nonetheless. This being said, if "embodiment" implies having a physical body, then I would argue that it is not a necessary requirement for intelligence. "Situatedness", being able to take (virtual or real) "actions" that influence the "inputs", may be.
  •  
    @Paul While I completely agree about the "embodiment done wrong" (or at least "not exactly correct") part, what you say goes exactly against one of the major claims which are connected with the notion of embodiment (google for "representational bottleneck"). The fact is your brain does *not* have resources to deal with big data. The idea therefore is that it is the body what helps to deal with what to a computer scientist appears like "big data". Understanding how this happens is key. Whether it is the problem of scale or of actually understanding what happens should be quite conclusively shown by the outcomes of the Blue Brain project.
  •  
    Wouldn't one expect that to produce consciousness (even in a lower form) an approach resembling that of nature would be essential? All animals grow from a very simple initial state (just a few cells) and have only a very limited number of sensors AND processing units. This would allow for a fairly simple way to create simple neural networks and to start up stable neural excitation patterns. Over time as complexity of the body (sensors, processors, actuators) increases the system should be able to adapt in a continuous manner and increase its degree of self-awareness and consciousness. On the other hand, building a simulated brain that resembles (parts of) the human one in its final state seems to me like taking a person who is just dead and trying to restart the brain by means of electric shocks.
  •  
    Actually on a neuronal level all information gets processed. Not all of it makes it into "conscious" processing or attention. Whatever makes it into conscious processing is a highly reduced representation of the data you get. However that doesn't get lost. Basic, low processed data forms the basis of proprioception and reflexes. Every step you take is a macro command your brain issues to the intricate sensory-motor system that puts your legs in motion by actuating every muscle and correcting every step deviation from its desired trajectory using the complicated system of nerve endings and motor commands. Reflexes which were build over the years, as those massive amounts of data slowly get integrated into the nervous system and the the incipient parts of the brain. But without all those sensors scattered throughout the body, all the little inputs in massive amounts that slowly get filtered through, you would not be able to experience your body, and experience the world. Every concept that you conjure up from your mind is a sort of loose association of your sensorimotor input. How can a robot understand the concept of a strawberry if all it can perceive of it is its shape and color and maybe the sound that it makes as it gets squished? How can you understand the "abstract" notion of strawberry without the incredibly sensible tactile feel, without the act of ripping off the stem, without the motor action of taking it to our mouths, without its texture and taste? When we as humans summon the strawberry thought, all of these concepts and ideas converge (distributed throughout the neurons in our minds) to form this abstract concept formed out of all of these many many correlations. A robot with no touch, no taste, no delicate articulate motions, no "serious" way to interact with and perceive its environment, no massive flow of information from which to chose and and reduce, will never attain human level intelligence. That's point 1. Point 2 is that mere pattern recogn
  •  
    All information *that gets processed* gets processed but now we arrived at a tautology. The whole problem is ultimately nobody knows what gets processed (not to mention how). In fact an absolute statement "all information" gets processed is very easy to dismiss because the characteristics of our sensors are such that a lot of information is filtered out already at the input level (e.g. eyes). I'm not saying it's not a valid and even interesting assumption, but it's still just an assumption and the next step is to explore scientifically where it leads you. And until you show its superiority experimentally it's as good as all other alternative assumptions you can make. I only wanted to point out is that "more processing" is not exactly compatible with some of the fundamental assumptions of the embodiment. I recommend Wilson, 2002 as a crash course.
  •  
    These deal with different things in human intelligence. One is the depth of the intelligence (how much of the bigger picture can you see, how abstract can you form concept and ideas), another is the breadth of the intelligence (how well can you actually generalize, how encompassing those concepts are and what is the level of detail in which you perceive all the information you have) and another is the relevance of the information (this is where the embodiment comes in. What you do is to a purpose, tied into the environment and ultimately linked to survival). As far as I see it, these form the pillars of human intelligence, and of the intelligence of biological beings. They are quite contradictory to each other mainly due to physical constraints (such as for example energy usage, and training time). "More processing" is not exactly compatible with some aspects of embodiment, but it is important for human level intelligence. Embodiment is necessary for establishing an environmental context of actions, a constraint space if you will, failure of human minds (i.e. schizophrenia) is ultimately a failure of perceived embodiment. What we do know is that we perform a lot of compression and a lot of integration on a lot of data in an environmental coupling. Imo, take any of these parts out, and you cannot attain human+ intelligence. Vary the quantities and you'll obtain different manifestations of intelligence, from cockroach to cat to google to random quake bot. Increase them all beyond human levels and you're on your way towards the singularity.
Luís F. Simões

Speeding swarms of sensor robots - 2 views

  • the algorithm is designed for robots that will be monitoring an environment for long periods of time, tracing the same routes over and over. It assumes that the data of interest — temperature, the concentration of chemicals, the presence of organisms — fluctuate at different rates in different parts of the environment.
  • But it turns out to be a monstrously complex calculation. “It’s very hard to come up with a mathematical proof that you can really optimize the acquired knowledge,”
  • The new algorithm then determines a trajectory for the sensor that will maximize the amount of data it collects in high-priority regions, without neglecting lower-priority regions.
  • ...1 more annotation...
  • At the moment, the algorithm depends on either some antecedent estimate of rates of change for an environment or researchers’ prioritization of regions. But in principle, a robotic sensor should be able to deduce rates of change from its own measurements, and the MIT researchers are currently working to modify the algorithm so that it can revise its own computations in light of new evidence. “
  •  
    smart!
1 - 7 of 7
Showing 20 items per page