1. Attachments are working again! Check out this thread for more details and to report any other bugs.

How/What Does the Prius' CPU(s) "Learn?"

Discussion in 'Gen 2 Prius Technical Discussion' started by Rokeby, Apr 7, 2008.

  1. Rokeby

    Rokeby Member

    Joined:
    Jan 21, 2008
    3,033
    708
    75
    Location:
    Ballamer, Merlin
    Vehicle:
    2008 Prius
    In a number of threads it has been reported that as you drive your Prius it
    "learns" something over time. Also, if for whatever reason the 12v battery
    is disconnected, the CPU(s) memory is at least diminished, if not outright
    erased. Afterwards the CPU(s) has to "relearn" what it formerly knew. This
    it has been said is evidenced by the FE/MPG dropping for a short period, up
    to a few tanks of gas.

    I'm getting decent FE, +/- 50 MPG and rising, no complaints. The CPU(s) and
    I seem to be learning well.

    First question; Just what is it that the CPU(s) "learn?" Is it about my style of
    driving, something like, "Goes easy on the blue bars, don't have to keep the
    HV battery at a high six bars." If you've got GPS, is it the terrain and your
    typical routes? (This would be very interesting and perhaps even downright
    scary.)

    Perhaps households with two Prius can help here. If the two drivers have
    vastly different driving styles, does each get noticeably different FE/MPG
    when you drive the "other car." (Yes I know, a highly unlikely event.)

    Second question: Does the car act or seem obviously different? Like maybe
    it's a little confused or hesitant in some way?
    -------------------------------------------------------------------------------------

    Summary and consolidation of postings through April 12, '08 appear in
    Posting #17.
     
  2. kocho

    kocho Member

    Joined:
    Oct 16, 2007
    134
    2
    0
    Location:
    Maryland
    Vehicle:
    2008 Prius
    Model:
    N/A
    I highly doubt the CPU learns my driving habits. Could "Learning" be a misconseption based in the fact that the CPU needs to go thru several driving cycles and specific driving conditions to complete all sensor and checks inspection cycles.

    But someone might surprise me with a fact how learning happens - the above is just my guess.

    I see not difference in driving dynamics before and after a DTC reset. Not sure if this counts as a CPU recet...

    I know that certain components do adjust to driving style, but that is based on immediate data while driving. All is forgotten moments later, when conditions change. For instance, a transmission might shift more aggressively if you floor the gas pedal. In my Camry, the A/C cuts off at full acceleration to add a few HP to the wheels where it is presumably needed. VSC also reacts to your driving as does speed-sensitive steering. Not sure if anything else does, besides a "memory" driver seat or steering wheel -;)

     
  3. dogfriend

    dogfriend Human - Animal Hybrid

    Joined:
    Feb 26, 2007
    7,512
    1,185
    0
    Location:
    Carmichael, CA
    Vehicle:
    2007 Prius
    I have experienced this phenomenon when I disconnected the 12V power so I could remove the seat without worrying about the Side Airbag going off. It took a couple of tankfuls before the car went back to "normal"

    If you lose backup power for the Engine ECU, then I would believe that you would lose the Short Term and Long Term Fuel Trim adjustments. This is data that all OBD 2 systems store to adjust for differences in individual engines.

    Also, there is a section in the Tech Training manual that says that the HV ECU stores "Historical Data" to refine the calculations used for the Energy Screen. I am attaching a screen shot of the page.
     

    Attached Files:

    3 people like this.
  4. Rokeby

    Rokeby Member

    Joined:
    Jan 21, 2008
    3,033
    708
    75
    Location:
    Ballamer, Merlin
    Vehicle:
    2008 Prius
    dogfriend,

    thank you for the Toyota Tech Manual cite. It is so relevant to this
    discussion that I am briniging it directly into the thread:

    --------------------------------------------------------------

    "Energy Monitor

    "The Energy Monitor, which includes a historical bar graph and total trip
    economy (MPG), is very accurate. Multiple, comparative calculations are
    performed by several computers.

    "Fuel useage and fuel economy are calculated by monitoring fuel injector
    duration and operating frequency. The ECU compares these values with
    miles traveled to calculate miles per gallon.

    "The battery ECU closely monitors energy consumption in Watts. By
    calculating the amount of energy spent, recovered and stored, the
    computer can calculate the required fuel burn. Fuel required to create
    this amount of energy is compared against the engine ECU fuel injection
    calculation to ensure accuracy.

    "Driving pattern, speed, and load characteristics are stored in the HV ECU
    as 'Historical Data.' Historical Data is used to further refine the MPG
    calculation. This data takes about three to six weeks to accumulate if the
    battery is disconnnected or the HV ECU is replaced."

    ---------------------------------------------------------------------------------

    The last paragraph would seem to explain the oft reported effect of 12V
    battery removal on reported MPG.

    Of significance to this thread is the question; Just what does Toyota
    mean by "driving pattern" and "load characteristics?"

    dogfriend, can you find out when the Tech Training Manual was written
    and/or updated?

    I have a suspicion that it was some while ago, say about 2005.
     
  5. Jack66

    Jack66 Kinda Jovial Member

    Joined:
    Jan 26, 2008
    766
    6
    0
    Location:
    Stafford, Virginia
    Vehicle:
    2008 Prius
    I've had the same question, but from the tech manual it seems to imply that the only use of the "learned" data is to make the MFD energy/MPG numbers more accurate?
     
  6. dogfriend

    dogfriend Human - Animal Hybrid

    Joined:
    Feb 26, 2007
    7,512
    1,185
    0
    Location:
    Carmichael, CA
    Vehicle:
    2007 Prius
    No, I don't know exactly when the tech training manual(s) were written, but they cover both the NHW11 and NHW20, so it must be 2004 or newer. You can download it here:

    AUTOMOTIVE TECHNICAL ARTICLES


    My impression from disconnecting the 12V battery two different times (once to study the construction of the seat, and once to actually modify it) is that the mileage dropped as reported on the MF display and miles driven per fillup. I have data of all fillups, but I need to reconstruct my spreadsheet after the file was lost, so I can't give exact figures at this time.

    Of course, YMMV. ;)
     
  7. qbee42

    qbee42 My other car is a boat

    Joined:
    Mar 2, 2006
    18,058
    3,073
    7
    Location:
    Northern Michigan
    Vehicle:
    2006 Prius
    Most if not all new electronic control systems on automobiles collect some information that is used for control system tuning. Learning is a reasonable description of the process. Much of the information compensates for minor differences from engine to engine. Yours may have more carbon on the plugs, or perhaps run in a colder climate. Some of the systems also adjust for driving styles.

    At present, the Prius has no knowledge of terrain. Work is being done by some researchers to use GPS and topographic data for optimizing the use of the battery, but as far as I know this has not been done on any production vehicle.

    Tom
     
  8. Jack66

    Jack66 Kinda Jovial Member

    Joined:
    Jan 26, 2008
    766
    6
    0
    Location:
    Stafford, Virginia
    Vehicle:
    2008 Prius
    Thanks, Dogfriend. I still wonder if your actual mileage and FE are effected by the CPU memory loss or if the data displayed on the MFD is not as accurate because it doesn't remember how to adjust the numbers to make them more accurate. I'm sure that the CPU makes some adjustments based on current engine sensor readings but the tech manual gives the impression that the history is used to make the numbers displayed in the MFD more accurate.
     
  9. dogfriend

    dogfriend Human - Animal Hybrid

    Joined:
    Feb 26, 2007
    7,512
    1,185
    0
    Location:
    Carmichael, CA
    Vehicle:
    2007 Prius
    Hobbit also has a paragraph on this in his page on rally hints:

    http://techno-fandom.org/~hobbit/cars/prius-rally-hints.txt

     
    Silver Pine Mica likes this.
  10. racerbob

    racerbob Member

    Joined:
    Mar 16, 2008
    138
    6
    0
    Location:
    S Florida
    Vehicle:
    2008 Prius
    Long term fuel trim is a set of "cells" in the engine ECU that store adjusted values of fuel mixture control. These are adjusted according to engine characteristics such as age, pumping ability (ring & valve condition, vacuum leaks, air filter restriction), cam wear, etc. When 12v battery is disconnected these may take a couple of days to fill up all the cells (blocks) with useful info. This is a trick that can work to your advantage sometimes in non hybrids when running for 1/4 mile times as the leaner, learned values are erased and replaced with default (usually richer) fuel values.:rolleyes:

    see: Block Learn Multiplier and Integrator
    for a more in-depth explanation.
     
    Silver Pine Mica likes this.
  11. Jack66

    Jack66 Kinda Jovial Member

    Joined:
    Jan 26, 2008
    766
    6
    0
    Location:
    Stafford, Virginia
    Vehicle:
    2008 Prius
    Someone asked me today if you could save the values stored in memory before disconnecting the battery then put them back after the battery is reconnected. Is that even possible? If it is volatile memory you should be able to write to it so it sounds reasonable.
     
  12. qbee42

    qbee42 My other car is a boat

    Joined:
    Mar 2, 2006
    18,058
    3,073
    7
    Location:
    Northern Michigan
    Vehicle:
    2006 Prius
    It's only possible if you are given read/write access to the memory. I am not aware of any way to do this, otherwise it should be a common service practice.

    If your friend is just replacing a battery, he can maintain 12 Volts on the system while he swaps out the old battery. Connect the maintaining source to the jump points under the hood.

    Tom
     
  13. patsparks

    patsparks An Aussie perspective

    Joined:
    Jul 12, 2007
    10,664
    567
    0
    Location:
    Adelaide South Australia
    Vehicle:
    2004 Prius
    Model:
    N/A
    Most ECUs with a knock sensor will remember the advance curve that is optimum for the fuel and the condition of the engine so they don't have the engine knocking all the time.
     
  14. kocho

    kocho Member

    Joined:
    Oct 16, 2007
    134
    2
    0
    Location:
    Maryland
    Vehicle:
    2008 Prius
    Model:
    N/A
    So, ...

    ... it seems the prius does *not* learn the driver's driving habits and adjust driveability to match them. Any info to the opposite?
     
  15. Rokeby

    Rokeby Member

    Joined:
    Jan 21, 2008
    3,033
    708
    75
    Location:
    Ballamer, Merlin
    Vehicle:
    2008 Prius
    Kudos to the many posters. You've brought up some most interesting
    points.

    From what's been said:
    * it is highly unlikely that terrain and location data is stored as "Historical
    Data."
    * there are ICE/FE related data that change over time that are
    necesssary for efficient ICE mangement and might be stored as Historical
    Data.

    The Tech Manual says that recovery of some of the data stored as
    Historical Data can take 3 to 6 weeks to recover/replace.
    It is hard to believe that such a long capture window is needed for ICE
    specific data. I have read through the Prius specific sections of the Tech
    Manual and it is perfectly clear that the CPU(s) are essentially
    continuously monitoring conditions that affect the ICE. In the case of fuel
    injector timing there are two correction regimes; short and long term. But
    the time intervals may be 2 and 10 minutes respectively.

    It seems to me that greatest variables affecting ICE operation are all
    external to the ICE itself; the operator, ambient temperature, and load
    carried in the vehicle. The CPU would have to make adjustments to the
    latter two rather quickly to run cleanly and efficiently. This would seem to
    indicate that the three week capture window is for operator related data.
    But I found nothing in the Tech Manual that would support this idea.

    In the past it took the early adopters great amounts of time to gather the
    data needed to identify and then validate data that we now take for
    granted: ICE stage transitions, stealth mode precursors, etc. So it would
    not surprise me that it will take a while to identify "What the CPU(s)
    learn."

    It is most probable that the process will outlive this thread. My hope is
    that we can make a first attempt at identifying data that the CPU(s)
    might need to "learn" and store away, including tentatively identifying the
    data that takes 3-6 weeks to collect.

    For my part, I'm still having difficulty coming to grips with one of the
    paragraphs extracted from the Tech manual:

    "The battery ECU closely monitors energy consumption in Watts. By
    calculating the amount of energy spent, recovered and stored, the
    computer can calculate the required fuel burn. Fuel required to create
    this amount of energy is compared against the engine ECU fuel injection
    calculation to ensure accuracy."

    It seems to me that regen energy, as it can come from hills and such is a
    wild card and would make this form of determining fuel use equivilance
    highly suspect.

    I clearly have a lot to learn.
     
  16. racerbob

    racerbob Member

    Joined:
    Mar 16, 2008
    138
    6
    0
    Location:
    S Florida
    Vehicle:
    2008 Prius
    "Long Term" to an ECU is only maybe every 2-5 seconds. STFT (short term fuel trim) is continuous and once it needs to deviate much more than 3% (plus or minus) from the programmed tables, it is stored as a LTFT correction and the short term reverts to a "new" zero point. LTFT may take several days to populate after a clear has occured (losing battery power or a ECU "reset" done with a scan tool). This is because many "blocks" must be populated. The only way this will happen is if the engine is cruising in every different rpm and MAP values that define the block tables. Normally the only driver specific habits that influence any stored values were related to transmission shift points. A Prius has a CVT so no shift points are necessary. It does however make sense that the Fuel Mileage Calculation could benefit by long term corrections that would get more accurate with a bigger database of information.
     
    Silver Pine Mica and olerox like this.
  17. Rokeby

    Rokeby Member

    Joined:
    Jan 21, 2008
    3,033
    708
    75
    Location:
    Ballamer, Merlin
    Vehicle:
    2008 Prius
    This summarizes and consolidates what posters have said so far.

    In summary, data that the Prius' CPU(s) might collect and save as "Historical
    Data" include
    :
    1. Short Term and Long Term Fuel Trim adjustments.
    2. Information that is used for control system tuning compensating for minor
    differences from engine to engine.
    3. Most ECUs with a knock sensor remember the optimum advance curve

    Further, "...the Prius has no knowledge of terrain." qbee 42

    The above three summary points come directly from these posts:

    dogfriend first pointed out: "Short Term and Long Term Fuel Trim
    adjustments. This is data that all OBD 2 systems store to adjust for
    differences in individual engines."
    (Post #3)

    In this same vein, racerbob said: "Long term fuel trim is a set of
    "cells" in the engine ECU that store adjusted values of fuel mixture control.
    These are adjusted according to engine characteristics such as age, pumping
    ability (ring & valve condition, vacuum leaks, air filter restriction), cam
    wear, etc. When 12v battery is disconnected these may take a couple of
    days to fill up all the cells (blocks) with useful info."
    (Post #10)

    And racerbob further added: "'Long Term' to an ECU is only maybe
    every 2-5 seconds. STFT (short term fuel trim) is continuous and once it
    needs to deviate much more than 3% (plus or minus) from the programmed
    tables, it is stored as a LTFT correction and the short term reverts to a 'new'
    zero point. LTFT may take several days to populate after a clear has occured
    (losing battery power or a ECU 'reset' done with a scan tool). This is because
    many "blocks" must be populated. The only way this will happen is if the
    engine is cruising in every different rpm and MAP values that define the
    block tables. Normally the only driver specific habits that influence any
    stored values were related to transmission shift points. A Prius has a CVT so
    no shift points are necessary. It does however make sense that the Fuel
    Mileage Calculation could benefit by long term corrections that would get
    more accurate with a bigger database of information."
    (Post #16)

    qbee42 noted: "Most if not all new electronic control systems on
    automobiles collect some information that is used for control system
    tuning... Much of the information compensates for minor differences from
    engine to engine. Yours may have more carbon on the plugs, or perhaps run
    in a colder climate."
    (Post #7)

    patsparks chimed in: "Most ECUs with a knock sensor will
    remember the advance curve that is optimum for the fuel and the condition
    of the engine so they don't have the engine knocking all the time."
    (Post #13)
     
    1 person likes this.
  18. CTitanic

    CTitanic Member

    Joined:
    Jan 23, 2011
    153
    21
    0
    Location:
    South Florida, FL, US
    Vehicle:
    2013 Prius
    Model:
    Three
    The only thing that page in that PDF is saying is that a historical data is saved and used to refine the MPG calculation.

    The whole learning theory sounds like science fiction. I haven't found anywhere any evidence of this technology being applied in a Prius. Is it possible? From the technical point of view it's possible. For example, Microsoft handwriting recognition feature is capable of learning from your handwriting but if multiple users are using the same profile instead of an improvement in the recognition process you will get exactly the contrary. And that's the scenario that already somebody described here in this thread. In very few occasions our Prius are driven by just one driver so to implement such technology in it we need profiles/accounts and I don't see this anywhere in my Prius or any other Prius yet.

    So what? The historical data will be saved again and unless you have a bipolar disorder the end calculations of this new set of data will mimic the previous results stored before the reset. Oh, but don't let anyone else drive your Prius during the "learning period".

    Do you know how many times your Prius was taken for a test drive before you bought it? ;) I should have asked to erase the old historical data before I left the dealer. :)
     
  19. dolj

    dolj Senior Member

    Joined:
    May 14, 2012
    7,487
    3,763
    0
    Location:
    Wellington, New Zealand
    Vehicle:
    2007 Prius
    Model:
    N/A
    Whether it sounds like science fiction or not, from my personal experience, it would indicate that there is something to this. And if you look at the MS handwriting recognition analogy you used, in reference to multiple drivers, you may well be onto something.

    I have twice (on two different Prii) had to disconnect the battery, once on one Prius to replace the 12V battery, and once on the second Prius the 12V went completely flat and so lost all car settings. On both occasions within a few weeks I started to see a marked improvement in fuel efficiency compared to what I was able to do before. On both cars, through circumstances I became the exclusive driver, but not exactly at that point. Prior my wife also shared driving duties.

    On one Prius I struggled to achieve better than 38.6 - 41.3 MPG (6.1 - 5.7 l/100km; 46.3 - 54.3 MPIG) and then after was able to see 44.3 - 54.7 MPG (5.3 - 4.3 l/100km; 53.2 - 65.7 MPIG). I don't know what specifically to attribute the improvement to, as the reset coincided with installing a new battery, and, I had recently changed the front tyres to not specifically LLR tyres, but lower RR than before - Yokohama C.Drive 2. I know I deliberately did not change my driving style, (and I know my wife did not either!), as I was keen to see if the new battery and tyres made any difference. Furthermore, I did continue to see continual improvement in FE, taking into account seasonal variation. Because of these various changes, I always wondered what exactly made the difference battery or tyres, or both.

    On the second Prius, the only thing that changed was the 12V being disconnected. I saw similar improvements after about 2-3 weeks. Since then I have changed the rear tyres to Goodyear Assurance Triple Max (from GY Assurance) and have seen a small improvement there too. My Consumption was around 45.2 - 47.0 MPG (5.2 - 5.0 l/100km; 54.3 - 56.5 MPIG) and now it is in the range of 52.3 - 56.0 MPG (4.5 - 4.2 l/100km; 62.8 - 67.3 MPIG). I'd like to point out that initially my wife drove this vehicle, but since I have driven it exclusively (in the last 3 or 4 tanks) I have been able to continually improve FE. This is in part to our different driving styles and I do focus on improving FE more than her, so that is definitely part of it, but, I have not done anything particularly different and the last 2 tanks and still have seen an improvement in FE.

    I will be interested to observe what happens when my wife can again share driving duties.



    You sound incredulous, but I do not think it is so far fetched. Yes the historical data will be save again. You may or may not mimic the previous results. If you are focused on continual improvement (and not necessarily be bipolar), presumably the data stored will be different. I would especially be leaning toward your last statement, even though you meant it in a sarcastic way.

    You may well be being facetious, but I have had those exact same thoughts, and after my experiences, would say you should do exactly that. If you do not have any prior hybrid driving experience and have learned to use any particular hybrid driving techniques, I'd go as far as to say wait year or more then reset the ECU. In saying that, am I seeing that the car is continually monitoring and updating to adapt to new conditions?

    Has anybody else made similar observations?
     
  20. dogfriend

    dogfriend Human - Animal Hybrid

    Joined:
    Feb 26, 2007
    7,512
    1,185
    0
    Location:
    Carmichael, CA
    Vehicle:
    2007 Prius
    You are trying to start an argument about a post from almost 6 years ago? You need to find a better hobby.
     
    uart likes this.