Playing with the Player Project

The Player Project is a robot-control software framework for interfacing with PC-based robots. Learn how to use it to interface with sensors, actuators and even full research robots.
Playing with Player

So far, our robot is awake, alert and ready to be told to do something interesting. Let's give it something to do. The CoroBot robot comes with a number of sensors and actuators—probably the easiest of which to interface with are the front- and rear-facing infrared ranging sensors and the mobility base's drive motors. Thus, we can write a small C program to talk to the Player server, read the IR sensors and drive the robot until it is 10cm away from an obstacle in front of the robot.

The first thing we have to do to interface with the Player server is open up a connection to it. For the sake of brevity, we will skip a lot of error checking, but you can download the full version of the code from the LJ FTP server (see Resources).

This code defines the variables we will use to talk to the Player server and the device interfaces in which we are interested:

#include "libplayerc/playerc.h"
static playerc_client_t*      clientHandle;
static playerc_position2d_t*  positionProxy;
static playerc_ir_t*          irProxy;

The clientHandle is used for talking to the Player server itself. The second position2d interface talks to the position2d interface, providing us with encoder information about how the wheels are moving and allowing us to send motor commands to the robot. We'll ignore the encoder information for this example. Lastly, the IR interface gives us information about the distances that the robot's IR sensors are reporting.

The next code snippet uses these proxies to interface with the server and these devices:

playerc_client_connect(clientHandle);

// convert our interface to a PULL interface,
// only updates when we read
playerc_client_datamode(clientHandle, PLAYER_DATAMODE_PULL);

// tell the robot to drop older messages
playerc_client_set_replace_rule(
       clientHandle, -1, -1, PLAYER_MSGTYPE_DATA, -1, 1);

// create the position proxy (controls the motors)
positionProxy = playerc_position2d_create(clientHandle, 0);
playerc_position2d_subscribe(positionProxy, PLAYER_OPEN_MODE);

// create the IR proxy (controls the IRs)
irProxy = playerc_ir_create(clientHandle, 0);
playerc_ir_subscribe(irProxy, PLAYER_OPEN_MODE);

We start off by connecting to the Player server and configuring our connection. We want to get new messages from the server only when we are ready for them, so we configure the connection for a pull-type arrangement. And, because we want only the most recent information (we don't care what the IR sensors were indicating a second ago, we care about what they are saying right now), we tell the server to report only the most recent data. If we really wanted, we could let Player ensure that every IR message was delivered, but that might result in getting less-than-fresh data and possibly driving into a wall.

After our connection is configured, we open up the position2d interface on the Player server and subscribe to it. Then, we do the same with the IR interface. So far, so good. Now we need to get the state of the IRs from the robot and tell it how to move the motors:


while (!timeToQuit) {
  // attempt to read from the client
  if (playerc_client_read(clientHandle) == 0)
    continue;      // nothing to read, try again.

  // read the IR distances and verify we have good data
  if (irProxy->data.ranges_count == 2) {
    frontIr = irProxy->data.ranges[0];
    rearIr  = irProxy->data.ranges[1];
  }

  // figure out how to drive
  runController(frontIr, rearIr,
                &desiredTranslation,&desiredRotation);

  playerc_position2d_set_cmd_vel(
      positionProxy, desiredTranslation,
      0, desiredRotation, 1);
}

Each time through the loop, we try to read the newest data from the robot. After a little sanity checking, we take the ranges reported by the IR sensors and feed them into a controller function. This controller does some magic processing (we'll talk about that later) and returns information on how we should drive the robot. Finally, we pass these driving commands back to the Player server and start it all over again.

All that's left now is to provide a runController function that maps from IR sensor readings to drive commands. The CoroBot driver accepts numbers in the range of –1.0 to +1.0 to tell how to drive the robot forward and backward: +1.0 means 100% power forward, –1.0 means 100% power in reverse, and 0.0 means stop. It accepts the same range for telling the robot how to turn: –1.0 means turn full power left, +1.0 means turn full power right, and 0.0 means drive straight ahead. Noting that the IR readings are provided in meters, we can use the following P-controller to drive our robot forward until we are 10cm away from a front obstacle. We even get a bonus for free—if we are closer than 10cm away, the robot will back up a bit until it is at the proper distance:

void runController(double frontIr, double rearIr,
                   double *translation,double *rotation)
{
  // convert our IR readings into drive commands
  *translation = (frontIr-0.1) * 3.0;
  *translation = *translation > 0.9? 0.9: *translation;
  *rotation    = 0.0;
}

And finally, good programmers always shut down their server connections when they are done:

void shutDownProxies()
{
  // close down proxies we have opened
  playerc_ir_unsubscribe(irProxy);
  playerc_ir_destroy(irProxy);
  playerc_position2d_unsubscribe(positionProxy);
  playerc_position2d_destroy(positionProxy);
  playerc_client_disconnect(clientHandle);
  playerc_client_destroy(clientHandle);
}

Building on the design we showed earlier, we can see how our drive-by-IR program interacts with the Player infrastructure. The CoroBot configuration file loads the phidgetIFK driver, which exposes an aio:0 device. This device allows the CoroBot driver to read the robot's onboard infrared sensors. The CoroBot driver also exposes the position2d and IR interfaces, which the drive-by-IR program reads with the help of the libplayerc library (Figure 5).

Figure 5. The Relationship between Several Devices and Interfaces When Using the Drive-by-IR Program

The Player Project offers a lot of functionality that there just isn't room to get into in one article. This includes robot simulation, support for numerous commercial robots of many different prices and qualities, and support for a whole slew of readily available devices. Its plugin system even allows you to build your own drivers for new devices, either to support new hardware or to implement new experimental algorithms. Give it a try, and give your computer a chance to stretch its legs.

______________________

White Paper
Linux Management with Red Hat Satellite: Measuring Business Impact and ROI

Linux has become a key foundation for supporting today's rapidly growing IT environments. Linux is being used to deploy business applications and databases, trading on its reputation as a low-cost operating environment. For many IT organizations, Linux is a mainstay for deploying Web servers and has evolved from handling basic file, print, and utility workloads to running mission-critical applications and databases, physically, virtually, and in the cloud. As Linux grows in importance in terms of value to the business, managing Linux environments to high standards of service quality — availability, security, and performance — becomes an essential requirement for business success.

Learn More

Sponsored by Red Hat

White Paper
Private PaaS for the Agile Enterprise

If you already use virtualized infrastructure, you are well on your way to leveraging the power of the cloud. Virtualization offers the promise of limitless resources, but how do you manage that scalability when your DevOps team doesn’t scale? In today’s hypercompetitive markets, fast results can make a difference between leading the pack vs. obsolescence. Organizations need more benefits from cloud computing than just raw resources. They need agility, flexibility, convenience, ROI, and control.

Stackato private Platform-as-a-Service technology from ActiveState extends your private cloud infrastructure by creating a private PaaS to provide on-demand availability, flexibility, control, and ultimately, faster time-to-market for your enterprise.

Learn More

Sponsored by ActiveState