CPS Summer School 2019

Participants in the CPS Summer School 2019 are invited to download and install the Acumen jar file for following the practice session on the second day (Tuesday). The main part that requires care is to make sure that you are running Java8.  Depending on your platform, can install the right version for Windows, Linux, or Mac. If one of these links does not work for you, get Version 8 from this site.

Once you have installed Java 8, download the Acumen distribution. To run Acumen, double click on the jar file in the top level directory.

We recommend that you carry out these steps before the first class (that is, before Monday) and post to the developers mailing list (acumen-developers@list.hh.se) on this post if you run into problems. If you complete the installation successful please post to confirm that all is well, and then familiarize yourself with the different examples that come with this distribution.

See you all on Monday!

Here is the model from the first day:

model Main(simulator) =

initially

 c1 = create Car((0,0,0),red),

 c2 = create Car((0,2,0),blue),

 x1=0, x1'=0, x1''=0,

 x2=0, x2'=0, x2''=0,

 z = 0

always

 if x1<20

  then x1'' = -(x1'-10)

  else if x1'>0

         then x1'' = -10

         else x1'' = 0,

 c1.pos = (x1,0,0),

 x2'' = -100*(x2-x1)-10*(x2'-x1'),

 c2.pos = (x2,2,0),

 z = x1-x2

model Car(pos,col) =

initially

 _3D = (),_Plot=()

always

 _3D = (Box center = pos+(0,0,0)

            color = 0.4*col + 0.6*white

            length = 4

            width = 1.55

            height = 1,

        Box center = pos+(0,0,1)

            color = 0.4*col + 0.6*white

            length = 2

            width = 1.55

            height = 1

        )

Mascots 2016

During the first two weeks of class, students taking the Cyber-Physical Systems (CPS) course at Halmstad University have a warm up exercise, namely, designing a mascot for their project team. The idea is to introduce them to Acumen's syntax and semantics, and to help them practice 3D geometry. As the video below shows, it is also a chance to show off some artistic talent!

Challenge Server

The Challenge Server is a new functionality that will enable Acumen users to interact by posing and responding to challenges. A challenge is a problem state in English, and the response is an Acumen model. A challenge must be posed with a template for the response and a grading model. The functionality is currently being testing, and should be available soon.

New Acumen Release (2014/12/1)

A new release of Acumen (2014/12/1) is now available.  This release features several enhancements, including switching to software rendering of 3D graphics, which makes the distributed executable fully functional on any computer that has Java installed. The latest semantics has more strict error checking, catching more inconsistencies your specification. Experimental support for matrix operations has also been included.

The following video showcases an example that comes with this distribution:

An updated user reference guide is available through the Help menu.  The examples have also been reorganized for easier access.  Essentially the same distribution was used successfully in the most recent edition of the Cyber-Physical Systems course taught at Halmstad University.

A New Breed of Acumen Ping Pong Players

WiffWaff (by Adam Duracz and Yingfu Zeng) is the Acumen player that won the first tournament of 3D Ping Pong project as part of the Halmstad CPS course in 2012. Terminator (by Jonas Jonson and Alexey Taktarov) won second place in first tournament in 2013. But local rankings can be deceiving! Terminator is a new bread of players that is much more sophisticated than anything developed in 2012. 

The result is clear from this game, where WiffWaff is on the left (red), and Terminator is on the right (black). Note the small balls that show Terminators predictions about bounce position and bat impact position. When a player turns into a yellow cube, it means it ran out of energy. Terminator is the first player that is able to successful complete the 40 second game against itself, returning every ball, and within the energy budget.

Congratulations to the new Alexey and Jonas!

A Preview of Acumen'13

A preview release of Acumen'12 is now available.  This release includes several experimental features, and improvements to the user interface.  More details to follow.

Postdoc position in virtual testing of advanced automotive safety systems

Halmstad University has an immediate opening for a postdoctoral associate in virtual testing of advanced automotive safety systems. Virtual testing involves extensive use of high-fidelity modeling and simulation methods.  The successful applicant will work within the project Next Generation Test Methods for Active Safety Functions, funded by Sweden’s Innovation Agency VINNOVA under the Strategic Vehicle Research Partnership FFI.

The description of work for the advertised position is to:

• Develop and experimentally validate a series of increasingly sophisticated mathematical models as relevant to the automated testing of future advanced driver assistance systems (ADAS). Aspects of interest include vehicle dynamics, sensors (GPS, radar, ladar), real-time communication services (V2V, V2I), road conditions, drivers, dependability, and test scenarios.
• Evaluate existing support for experimental validation of mathematical models using empirically gathered data and statistical methods
• Evaluate the convenience and fidelity of various modeling and simulation tools for performing virtual experiments during research and development of ADAS, in particular during the early stages of design and requirement specification.
• Document the results in a technical report valuable to the industrial project partners and establish a suitable basis for a journal paper on the scientific contributions. Depending on the pace of progress, publication in an appropriate journal.

The position is for one year, with a salary in the range of SEK 30’000 to 33’000 per month. Depending on the performance of the successful applicant, there will be opportunities for continued employment within upcoming projects. More details about the project can be found on here and further inquiries can be addressed to Roland Philippsen, Tony Larsson, Walid Taha, or Magnus Jonsson at firstname.lastname@hh.se.

To apply to the position, please send in all application materials as one PDF file attached to an email addressed to registrator@hh.se.  The title of the email must be “Application to NG-Test Postdoc (Position number 6/12)”.  The PDF file must consist of 1) a cover letter stating the reasons for interest in this position, and long term career goals, 2) CV or resume, 3) list of references, 4) examples of published or thesis work, and 5) Copy of PhD degree.

Applications will be processed on a first-come first serve basis. Priority will be given to applications arriving before September 15th, 2012.

A Model of a Ping Pong Game

Last year in the programming languages course at Halmstad University, students worked in groups to develop different parts of a ping pong game.  Later, Yingfu Zeng combined these projects into one model that streamlined these components and further developed below.  The result is what you find below in the rest of this post.

The model below was used in the first tournament in cyber physical systems course.  Out of seven entries, the winning entry was team Virtue by Victor Vasilev and Carlos Fuentes.  The winning entry was able to score 7.5 out of a maximum of 12 possible points.   The benchmark model, team WiffWaff by Adam Duracz and Yingfu Zeng, shows that it is possible to score 11 out of 12 points.

NEW!  Check out the videos and analysis!


/**
* Program:   3-D ping pong 
* Author :   Yingfu Zeng, Walid Taha
* Date   :   2012/02/11
* License:   BSD, GPL(V2), or other by agreement with Walid Taha
**/
class Ball ()
 private
  mode = "Fly";
  k_z  = [1,1,-0.99];       // Coefficient of restitution
  k2   = 1/6;               // Coefficient of the air resistance
  p    = [0,0,0.5];         // Position of the ball
  p'   = [5,1,-3];
  p''  = [0,0,0];
  _3D  = ["Sphere",[0,0,0.5],0.03,[1,1,1],[0,0,0]];
 end
 _3D [=] ["Sphere",p,0.03,[1,1,1],[0,0,0]];
 // Valid modes
 if mode ~= "Fly" && mode ~= "Bounce" && mode ~= "Freeze"
   mode = "Panic!";
 end;
 switch mode
  case "Fly"
   if dot(p,[0,0,1]) < 0 && dot(p',[0,0,1])< 0
    mode = "Bounce";
   else
    p'' [=] -k2 * norm(p') * p' + [0,0,-9.8];
   end;
  case "Bounce"
    p'   =  p' .* k_z;    // Bouncing  will lose some energy
    mode = "Fly";
  case "Freeze"           // The ball becomes red to show what is going wrong
    p'  [=] [0,0,0]; p'' [=] [0,0,0];
    _3D [=] ["Sphere",p,0.03,[1,0,0],[0,0,0]];
  case "Panic!"
  end
end




class BatActuator(p1)   
 private
  p       = p1;
  p'      = [0,0,0];
  angle   = [0,0,0];
  energy  = 0;
  energy' = 0;  
 end
  if norm(p') > 5
   p' = p'/norm(p') * 5 ;
  end;
  energy' [=] norm(p'); 
end




class Bat(n,p1)
 private
  p     = p1;
  p'    = [0,0,0];
  angle = [0,0,0.1];
  displayAngle = [0,0,0];
  mode  = "Run";
  _3D   = ["Cylinder",p1,[0.15,0.05],[0.1,0.1,0.1],[0,0,0.5]];
 end
 switch mode 
  case "Run"
   if n == 2
     displayAngle  [=] [0,dot(angle,[0,0,1])*(3.14/2)/norm(angle),
                      dot(angle,[0,1,0])*(3.14/2)/norm(angle)]+[0,0,3.14/2];
    _3D            [=] ["Cylinder",p+[0.05,0,0],[0.15,0.05],
                        [0.1,0.1,0.1],displayAngle];
   else
      displayAngle [=] [dot(angle,[0,0,1])*(3.14/2),0,
                    dot(angle,[0,1,0])*(3.14/2)]+[0,0,3.14/2];
     _3D           [=] ["Cylinder",p+[-0.05,0,0],[0.15,0.05],
                        [1,0.1,0.1],-1 * displayAngle];
 end;
  case "Rest"
    p'            [=] [0,0,0];
 _3D           [=] ["Box",p+[-0.05,0,0],[0.3,0.3,0.3],
                        [1,1,0.1],-1 * displayAngle];
 end
end




/**
*Position and velocity of ball(ballp,ballv) always provided estimately;
*Once player decides to hit the ball, change the hit variable to true,
*the Game class will notice and caculate the output velocity of the ball.
**/
class Player(n)
 private
  mode      = "Wait";
  bounced   = false;       // Tell whether the ball bounced or not
  serve = false;           // The Game class will set serve flag to true 
  hit   = false;           // when it's your turn
  count = 0;
  ballv = [0,0,0];
  ballp = [0,0,0];
  batp  = [1.6,0,0.2];
  v     = [0,0,0];         // Bat's speed 
  batAngle   = [0,0,0.1];  // Normal vector of the bat's plane
  batAngle'  = [0,0,0];
  // Player(1) starts at [-1.6,0,0.2], Player(2) starts at [1.6,0,0.2]
  startPoint = [1.6*(-1)^n,0,0.2]; 
  t   = 0;
  t'  = 1;
 end
 if mode ~= "Wait" && mode ~= "Prepare" && mode ~= "Hit"
   mode = "Panic!";
 end;
 t'  [=] 1;
 switch mode
  case "Wait"               // While waiting, moving the bat to starting point
   count      = 0;
   if n == 1 
     v         [=] startPoint-batp;
   else
     v         [=] startPoint + [0,0.75,0] - batp;
   end;
   batAngle' [=] [0,0,0]-batAngle;
   hit    = false;
   if serve == true
    mode    = "Prepare";
    bounced = false;
   else
    mode = "Wait";
   end;
  case "Prepare"             // Prepare to hit the ball
   if bounced == true        // After the ball has bounced,
                             // start moving the bat towards the ball
     v [=] (ballp-batp).*[0,20,0] + (ballp-batp).*[0,0,25] +
           (ballp+[0.12*(-1)^n,0,0]-batp).*[25,0,0];
     if norm(batp - ballp)<0.15 && abs(dot(ballp,[1,0,0])) >= 
                               abs(dot(startPoint,[1,0,0]))
      count = count+1;
      mode  = "Hit";
     end;
   end;
   // When the ball has bounced and it is at the highest position
   if count > 0 && dot(ballv,[0,0,1]) < 0.1 && bounced == true  
    mode = "Hit";     // This player decide to hit.
   end;
   if dot(ballp,[0,0,1]) < 0 && bounced == false
    bounced = true;
   end;
   if(serve ~= true)
     mode = "Wait";
   end;
 case "Hit"           // Decide how you want hit the ball, 
  if n == 2
   if(t<1||t>5)       // you may want to check the formulas 
                      // in the BallActuator() class
    v        = [-1.38,0.40,1.2];
    batAngle = [0.9471,0.25,-0.2];
   else
    if t > 4 && t < 5
     v        = [-0.88,-0.5,0.2];
     batAngle = [0.9471,0.25,-0.2];
 else
     v        = [-1.7,-0.2,3.86];
     batAngle = [0.96,-0.1,-0.2258];
 end;
   end;  
  else
   if(dot(ballv,[0,1,0]) < 0)
    v        = [0.1,-0.15,3.85];
    batAngle = [-0.938,-0.162,-0.29];
   else
    v        = [1,0,2.85];
    batAngle = [-0.938,0.202,-0.29];
    end;
   end;
  serve  = false;
  hit    = true;
  mode   = "Wait";
 case "Panic!"
 end
end




class Table()   // The table
 private
                // Table
 _3D = [["Box", [0,0,-0.05],[3,1.5,0.03],[0.1,0.1,1.0],[0,0,0]],
                // TableBases 1~4 
        ["Box", [-1.4,0.6,-0.3-0.04], [0.05,0.05,0.6], [0.8,0.8,0.8],[0,0,0]], 
        ["Box", [-1.4,-0.6,-0.3-0.04], [0.05,0.05,0.6], [0.8,0.8,0.8],[0,0,0]],
        ["Box", [1.4,-0.6,-0.3-0.04], [0.05,0.05,0.6], [0.8,0.8,0.8],[0,0,0]],
        ["Box", [1.4,0.6,-0.3-0.04], [0.05,0.05,0.6], [0.8,0.8,0.8],[0,0,0]],
          // Net
        ["Box", [0,0,0.125-0.02], [0.05,1.5,0.25], [0.2,0.8,0.2],[0,0,0]], 
                 // MiddleLine  
  ["Box", [0,0,0],[3,0.02,0.02-0.02],[1,1,1],[0,0,0]]]           
 end
end




class BallActuator()  // Calculate result of impact
  private
   mode="Initialize";
   v1 = [0,0,0];      // Input ball speed
   v2 = [0,0,0];      // Output ball speed
   v3 = [0,0,0];      // Bat's speed during the impact
   angle = [0,0,0];   // Bat's normal vector
   done  = false;
   action = 0;
 end
  if mode ~= "Initialize" && mode ~= "Calculate" && mode ~= "Wait"
   mode = "Panic!";
  end;
 switch mode
  case "Initialize"
   done[=]false;
   if action == 1
    mode = "Calculate";
   end;
 case "Calculate"
  v2     = v1-dot(2.*(v1-v3),angle)*angle;
  action = 0;
  if action == 0
   mode = "Wait";
  end;
 case "Wait"
  done [=] true;
 case "Panic!"
 end
end




// Sample the velocity of the ball and feed back to the players.
class BallObserver()  
 private
  mode = "Sample";
  p  = [0,0,0];
  v  = [0,0,0];
  pp = [0,0,0];
  ap = [0,0,0];
  t  = 0;
  t' = 1;
 end
 t'[=]1;
 if mode ~= "Sample" && mode ~= "Estimate0" && mode ~= "Estimate1"
  mode = "Panic!";
 end;
 switch mode
   case "Sample"
    if t > 0
     pp  = p;
     t   = 0;
     mode= "Estimate0"
    end;
   case "Estimate0"
 if t == 0.01   // Calculate the average speed
     ap   = p;
     mode = "Estimate1";
    end;
   case "Estimate1"
    v    = dot((ap-pp),[1,0,0])/0.01*[1,0,0]+dot((ap-pp),[0,0,1])/0.01*[0,0,1]+
        dot((ap-pp),[0,1,0])/0.01*[0,1,0];
    mode = "Sample";
    t    = 0;
   case "Panic!"
  end
end




class Referee()  // This class will monitors the whole process of the game.
 private
  mode="Initialize";
  x = 0;x' = 0;
  z = 0;z' = 0;
  y = 0;
  t = 0;t' = 1;
  player1Score = 0;
  player2Score = 0;
  serveNumber  = 2;
  lastHit      = 0;
  reason       = "Nothing";
  checked      = false;    // For the net checking
  bounced      = false;
  restart      = 0;        // Tell the Game to restart
  acknowledged = 0;        // Check if the Game class has received 
                           //  the restart signal
  bounceTime   = 0;
  status       = "Normal"
 end
 if mode ~= "Initialize" && mode ~= "Player1Lost" && mode ~= "Player2Lost" 
    && mode ~= "SendMessage" && status ~= "Normal" && reason ~= "Nothing"
    && status ~= "Report" && reason ~= "BallOutOfBoundary"
 && reason ~= "BallBouncedTwice" && reason ~= "BallTouchNet"
  mode = "Panic!";
 end;
  t'[=]1;
  if z<0.05 && z'<0 && status == "Normal"  // Check if anyone fouls
   if (abs(y)>0.78||abs(x)>1.53) && status == "Normal"
    reason     = "BallOutOfBoundary";
    if bounced == false
     if x>0
      mode = "Player1Lost";
     else
      mode = "Player2Lost";
     end;
    else
     if bounced == "YesIn2"    // The ball has bounced in player2's court, 
      mode = "Player2Lost"     // and out of boundary now, so player2 lose.
     end;
     if bounced == "YesIn1"
      mode = "Player1Lost";
     end;
    end;
    status = "Report";
   end;
   if(abs(y)<0.78 && abs(x)<1.53) && bounced ~= false  
      && t>(bounceTime+0.1) && status=="Normal"
 // The ball has bounced twice in player2's court  
    if bounced == "YesIn2" && x > 0 
     mode   = "Player2Lost";
     reason = "BallBouncedTwice";
bounceTime = t;
    end;
 // The ball has bounced twice in player1's court
    if bounced == "YesIn1" && x < 0 
     mode   = "Player1Lost";
     reason = "BallBouncedTwice";
bounceTime = t;
    end;
   end;
   if x<0 && x>-1.5 && bounced == false && status == "Normal"
    bounced    = "YesIn1";
    bounceTime = t;
   end;
   if x>=0 && x<1.5 && bounced == false && status == "Normal"
    bounced    = "YesIn2";
    bounceTime = t;
   end;
 end;




 if bounced == "YesIn1" && x>0 && status == "Normal"
  bounced = false
 end;
 if bounced == "YesIn2" && x<=0 && status == "Normal"
  bounced = false
 end;
  // Time to check if the ball touches the net
 if abs(x)<0.025 && t>0.1 && checked == false && status == "Normal"   
  if z<0.25
    if x'>0
     mode   = "Player1Lost";
    else
     mode   = "Player2Lost"
    end;
    reason  = "BallTouchNet";
    checked = true;
  end;
 end;
switch mode
 case "Initialize"
 case "Player1Lost"
  player2Score = player2Score+1;
  mode = "SendMessage";
 case "Player2Lost"
  player1Score = player1Score+1;
  mode = "SendMessage";
 case "SendMessage"
  t = 0; // Wait until the Game class gets the restart signal
  restart = 1;
  if acknowledged == 1
    mode = "Initialize";
    acknowledged = 0;
    restart = 0;
    status  = "Normal";
    checked = false;
    bounced = false;
  end;
  case "Panic!"
 end
end




/**
* The parent of all the other classes, who controls the
* whole process of the game.
**/
class Game ()
 private
  ball    = create Ball ();
  ballob  = create BallObserver();
  actuator= create BallActuator();
  batActuator1 = create BatActuator([-1.6,0,0.2]);
  batActuator2 = create BatActuator([1.6,0,0.2]);
  player1 = create Player(1);
  player2 = create Player(2);
  bat1    = create Bat(1,[-1.6,0,0.2]);
  bat2    = create Bat(2,[1.6,0,0.2]);
  table   = create Table();
  gameMonitor = create Referee();
  mode    = "Player2Serve";       // Player2 starts first
  player2Score = 0;
  player1Score = 0;
  serveNumber  = 2;
  t  = 0;
  t' = 1;
  maxEnergy    = 18;
 end
  if mode ~= "Restart" && mode ~= "Player1Serve" && mode ~= "Player2Serve" 
  && mode ~= "Impact"  && mode ~= "Freeze" && mode ~= "ChangeSide"
  && mode ~= "Act"
   mode = "Panic!"
  end;
  t'[=]1;
  gameMonitor.x  [=] dot(ball.p,[1,0,0]);
  gameMonitor.x' [=] dot(ball.p',[1,0,0]);
  gameMonitor.z  [=] dot(ball.p,[0,0,1]);
  gameMonitor.z' [=] dot(ball.p',[0,0,1]);
  gameMonitor.y  [=] dot(ball.p,[0,1,0]);
  gameMonitor.serveNumber [=] serveNumber;
  player1Score  [=] gameMonitor.player1Score;
  player2Score  [=] gameMonitor.player2Score;
  ballob.p          [=] ball.p;
  player1.ballp     [=] ballob.p;
  player2.ballp     [=] ballob.p;  
  player1.ballv     [=] ballob.v;
  player2.ballv     [=] ballob.v;
  if bat1.mode ~= "Rest"
   batActuator1.p' [=] player1.v;
  end;
  if bat2.mode ~= "Rest"
   batActuator2.p' [=] player2.v;
  end;
  player1.batp  [=] bat1.p;
  player2.batp  [=] bat2.p;
  batActuator1.angle [=] player1.batAngle;
  batActuator2.angle [=] player2.batAngle;
  bat1.p  [=] batActuator1.p;
  bat1.p' [=] batActuator1.p';
  bat2.p  [=] batActuator2.p;
  bat2.p' [=] batActuator2.p';
  bat1.angle [=] batActuator1.angle;
  bat2.angle [=] batActuator2.angle;
  if batActuator1.energy > maxEnergy
     bat1.mode = "Rest";
 bat1.p'   = [0,0,0];
 batActuator1.p' [=] [0,0,0];
  end;
  if batActuator2.energy > maxEnergy
     bat2.mode = "Rest";
 bat2.p'   = [0,0,0];
 batActuator2.p' [=] [0,0,0];
  end; 
 switch mode
  case "Restart" // Put everything back to the starting point
   ball.p            = [0,0,0.5];
   ball.p'           = [5,1,-3];
   bat2.p            = [1.6,0,0.2];
   player2.batp      = [1.6,0,0.2];
   player2.v         = [0,0,0];
   player2.batAngle  = [0.01,0,0];
   player2.bounced   = false;
   player2.ballp     = [1.6,0,0.2];
   bat1.p            = [-1.6,0,0.2];
   player1.batp      = [-1.6,0,0.2];
   player1.v         = [0,0,0];
   player1.batAngle  = [0.01,0,0];
   player1.bounced   = false;
   player1.ballp     = [-1.6,0,0.2];
   batActuator1.p    = [-1.6,0,0.2];
   batActuator2.p    = [1.6,0,0.2];
   serveNumber       = 2;
   gameMonitor.bounced      = false;
   gameMonitor.checked      = false;
   gameMonitor.acknowledged = 1;
   mode         = "Player2Serve";
   player1.mode = "Wait";
   player2.mode = "Wait";
  case "Player2Serve" // Player 2 is serving
   player1.serve [=] false;
   player2.serve [=]  true;
   if player2.hit == true && norm(bat2.p - ball.p) < 0.15
    mode = "Impact"
   end;
   if gameMonitor.restart == 1
    mode = "Freeze";
    t    = 0;
   end;
 case "Player1Serve" // Player 1 is serving
  player2.serve [=] false;
  player1.serve [=] true;
  if player1.hit == true && norm(bat1.p - ball.p) < 0.15
   mode = "Impact"
  end;
  if gameMonitor.restart == 1
   mode = "Freeze";
   t    = 0;
  end;
 case "Impact" // When one player hits the ball
  actuator.v1 = ball.p';
  if serveNumber == 2 // Give player2's data to actuator
   batActuator2.p' = player2.v;
   bat2.p'         = batActuator2.p';
   actuator.v3     = bat2.p';
   bat2.angle      = player2.batAngle;
   actuator.angle  = bat2.angle;
   actuator.action = 1; // Tell actuator to act
   gameMonitor.lastHit = 2;
   mode = "Act";
    if gameMonitor.restart == 1
      mode = "Freeze";
      t = 0;
    end;
  end;
  if serveNumber == 1 // Give player1's data to actuator
   batActuator1.p' = player1.v;
   bat1.p'         = batActuator1.p';
   actuator.v3     = bat1.p';
   bat1.angle      = player1.batAngle;
   actuator.angle  = bat1.angle;
   actuator.action = 1; // Tell actuator to act
   gameMonitor.lastHit = 1;
   mode = "Act";
   if gameMonitor.restart == 1
    mode = "Freeze";
    t    = 0;
   end;
  end
 case "Act" // Wait till actuator finish
  if gameMonitor.restart == 1
   mode = "Freeze";
   t    = 0;
  end;
  if actuator.done == true
   ball.p'       = actuator.v2;
   actuator.mode = "Initialize";
   mode          = "ChangeSide";
  end;
 case "ChangeSide" // Change the serve number
  if gameMonitor.restart == 1
   mode = "Freeze";
   t    = 0;
  end;
  if serveNumber == 2 && dot(ball.p,[1,0,0]) >0 && gameMonitor.restart ~= 1
   serveNumber     = 1;
   mode            = "Player1Serve";
   player1.mode    = "Wait";
   player1.bounced = false;
  end;
  if serveNumber == 1 && dot(ball.p,[1,0,0]) <= 0 && gameMonitor.restart ~= 1
   serveNumber     = 2;
   mode            = "Player2Serve";
   player2.mode    = "Wait";
   player2.bounced = false;
  end; 
 // When someone fouls, showing what's going wrong for 1 second
 case "Freeze"          
   if t<1
    ball.mode = "Freeze";
   else
    mode      = "Restart";
    ball.mode = "Fly";
   end;
 case "Panic!"
 end
end




class Main(simulator)
 private
  mode = "Initialize";
 end
switch mode
 case "Initialize"
  simulator.endTime = 20;
  create Game();
  mode = "Persist";
 case   "Persist"
 end
end

PhD and Research Programmer Openings at Rice and Texas A&M

New announcements have just been posted for four PhD positions and one research programer positions funded by a new NSF project lead by Professor Taha.  The positions are hosted by Rice and Texas A&M Universities, but participants will spend significant periods of time at Halmstad University.  Please forward these announcements to strong candidates.  Note that the two deadlines are as early as January 6th.

Research Programmer in Computer Science at Rice

Major technical challenges get in the way of developing novel cyber-physical systems (CPS). Overcoming these challenges can significantly accelerate the design process. A new NSF CPS project on Robot Design is building foundations and tools to achieve this goal for novel robotics applications. The project focuses on issues relating to modeling hybrid (continuous/discrete) systems and model validation. The project, which is a close collaboration between Computer Science and Mechanical Engineering at Rice University, Mechanical Engineering at Texas A&M, and Halmstad University in Sweden, will develop and validate foundations, methods, and tools for the modeling and simulation of hybrid and cyber-physical systems.

The project has an opening a research programmer with Professor Taha at Rice University. The successful candidates will work closely a team of four professors and four PhD students working on various aspects of this project, contributing and coordinating the development of the Acumen modeling language for the purposes of this project.

Applicants for this position must have strong skills and interest in programming language design and theory; functional programming; and test-driven development.  Excellent communication and collaboration skills are required as the student will be expected to work as part of a large, international research team investigating modeling languages. Priority will be given to applicants who already hold a doctoral degree in Computer Science or Mechanical Engineering, but applicants with suitable skills and interests who hold a doctorate in Mathematics, Physics, or Electrical Engineering are also encouraged to apply. Desirable experience includes working interval and enclosure methods, exact real arithmetic, domain theory, operational semantics, formal verification, differential equations, hybrid/impulsive differential equations, and mechanical hardware design and control, especially in the context of robotics. Desirable skills include facility with Scala, Java, Haskell, OCaml, ML, Scheme, Mathematica, Maple, OpenGL, Matlab, Simulink, the Open Dynamics Engine (ODE), or other modeling and simulation tools.

The position is for one year, renewable to two years, and is subject to the availability of funding. The successful applicant will spend part of their time in Halmstad, Sweden (50%) as part of the collaboration and must be ready to start the position by May or June 2012.

The deadline for applications is January 15th, 2012. Application for this position should be sent directly to Professor Taha by email. The title of the email should be (without quotes) "Application for Robot Design Research Programmer Position". The application should consist of one, self-contained PDF attachment. The application should include 1) a brief cover letter including motivation for the application, 2) a CV, and 3) a list of names of references. It is also suggested that the application include online pointers to previous publications or software produced by the applicant.

Doctoral (PhD) student in mechanical engineering at Texas A&M

Major technical challenges get in the way of developing novel cyber-physical systems (CPS). Overcoming these challenges can significantly accelerate the design process. A new NSF CPS project on Robot Design is building foundations and tools to achieve this goal for novel robotics applications. The project focuses on issues relating to modeling hybrid (continuous/discrete) systems and model validation. The project, which is a close collaboration between Computer Science and Mechanical Engineering at Rice University, Mechanical Engineering at Texas A&M, and Halmstad University in Sweden, will develop and validate foundations, methods, and tools for the modeling and simulation of hybrid and cyber-physical systems.

The project has an opening for a Mechanical Engineering doctoral (PhD) student with Professor Ames at the University of Texas A&M. The student's dissertation will focus modeling and simulation of lower-limb robots and will involve the evaluation and development of new technologies for modeling and simulation of such systems. The successful candidate will work closely with Computer Scientists developing new computational tools to support modeling and simulation.

Applicants for must have strong skills and interest in mechanics, control theory, rigid body dynamics, and robotics.  Excellent communication and collaboration skills are required as the student will be expected to work as part of a large, international research team investigating modeling languages. Priority will be given to applicants with a masters in Mechanical or Electrical Engineering, but students with suitable skills and interests who hold a masters in Mathematics, Physics, or Computer Science are also encouraged to apply. Desirable experience includes working differential equations, hybrid/impulsive differential equations, or with mechanical hardware design and control, especially in the context of robotics and upper limb rehabilitation robots. Desirable skills include facility with Mathematica, Maple, Matlab, Simulink, the Open Dynamics Engine (ODE), or other modeling and
simulation tools.

Selected students are will spend part of their time in Halmstad, Sweden (25%) as part of the collaboration and must be ready to start the position by May or June 2012.

The deadline for applications is March 1st. Application for these positions is done through the on-line application form. Your application (and in particular your Research Statement) must make it clear that  your are applying for a PhD position with Professor Ames on the Robot Design project.

Two Doctoral (PhD) students in Computer Science at Rice University

Major technical challenges get in the way of developing novel cyber-physical systems (CPS). Overcoming these challenges can significantly accelerate the design process. A new NSF
CPS project on Robot Design is building foundations and tools to achieve this goal for novel robotics applications. The project focuses on issues relating to modeling hybrid (continuous/discrete) systems and model validation. The project, which is a close collaboration between Computer Science and Mechanical Engineering at Rice University, Mechanical Engineering at Texas A&M, and Halmstad University in Sweden, will develop and validate foundations, methods, and tools for the modeling and simulation of hybrid and cyber-physical systems.

The project has an opening for two Computer Science doctoral (PhD) student with Professor Taha at Rice University. Each student's dissertation work will have distinct technical focus. One will focus on developing the meta-theory for relating user-level descriptions of mechanical systems, mathematical equations describing the behavior of such systems, and the simulation codes needed to solve these equations.  The other will focus on developing methods for the simulation of hybrid systems, including zero-crossing and Zeno behavior. The succesful candidates will work closely with Mechanical engineerings who are working on modeling and simulation of new experimental robots.  The engineers will use and evaluate the results of the work of the PhD students.

Applicants for this position must have strong skills and interest in programming language design and theory; functional programming; and test-driven development.  Excellent communication and collaboration skills are required as the student will be expected to work as part of a large, international research team investigating modeling languages. Priority will be given to applicants with a masters in Computer Science or Electrical Engineering, but students with suitable skills and interests who hold a masters in Mathematics, Physics, or Mechanical Engineering are also encouraged to apply. Desirable experience includes working with interval and enclosure methods, exact real arithmetic, domain theory, operational semantics, formal verification, differential equations, hybrid/impulsive differential equations, and mechanical hardware design and control, especially in the context of robotics. Desirable skills include facility with Scala, Java, Haskell, OCaml, ML, Scheme, Mathematica, Maple, OpenGL, Matlab, Simulink, the Open Dynamics Engine (ODE), or other modeling and simulation tools.

Selected students are will spend part of their time in Halmstad, Sweden (50%) as part of the collaboration and must be ready to start the position by May or June 2012.

The deadline for applications is January 6th, 2012 January 16th, 2012 (Rice Computer Science extended the deadline). Application for these positions is done through the on-line application form. Your application (and in particular your Research Statement) must make it clear that your are applying for a PhD position with Professor Taha on the Robot Design project.

Doctoral (PhD) student in mechanical engineering at Rice University

Major technical challenges impede the effectiveness of modeling and simulation of novel robot designs. The NSF CPS project on Robot Design develops foundations and tools for overcoming these challenges. The project focuses on issues relating to improving hybrid modeling formalisms demonstrating the physical validity of models developed in this formalism.  The project is a close collaboration between Computer Science and Mechanical Engineering at Rice University, Mechanical Engineering at Texas A&M, and Halmstad University in Sweden, will develop and validate foundations, methods, and tools for the modeling and simulation of cyber-physical systems (CPS).

The project has an opening for doctoral (PhD) student with the MAHI lab at Rice University. The student will contribute to the project by using a new modeling language to model and validate the MAHI lab's upper-extremity exoskeleton robot (the RiceWrist).

Applicants must strong skills in dynamic systems and controls, including modeling, simulation, and hardware implementation.  Excellent communication and collaboration skills are required as the student will be expected to work closely with software team building the modeling language. Priority will be given to applicants with a masters in Mechanical or Electrical Engineering. Desirable experience includes mechanical hardware design and control experience, especially in the context of robotics and upper limb rehabilitation robots. Desirable skills include Mathematica, Maple, Matlab, Simulink, other modeling and simulation tools.

The student wil spend part of their time in Halmstad, Sweden (25%) as part of the collaboration.

Students can apply to enter the program either in January 2012 (deadline November 1, 2011) or August 2012 (deadline February 1, 2012). Applicants desiring Fall 2012 start will be encouraged to begin research in May/June 2012.  Application for this position is done through the on-line application form.