Updating and Reconfiguring P2OS

PID Parameters

The P2OS configuration parameters include settings for the PID motors controls for translation and rotation of the robot. The translation PID values also apply to independent wheel velocities. The default values shown in the Table are for a moderately loaded robot. Experiment with different values to improve the performance of your robot in its current environment.

Note that with P2OS version 1.J and later, the PID values have changed from earlier versions due to changes in the drive servers. Overall drive performance has improved with the latter version and is much less sensitive individual PID values. Consequently, we recommend the listed values for all Pioneer 2 and PeopleBot robots, whereas earlier P2OS versions required much more tuning of the parameters depending on robot type and configuration/payload.

The P term value Kp increases the overall gain of the system by amplifying the position error. Large gains will have a tendency to overshoot the velocity goal; small gains will limit the overshoot but cause the system to become sluggish. We’ve found that a fully loaded robot works best with a Kp setting of around 15 to 20, whereas lightly loaded robot may work best with Kp in the range of 20 to 30.

The D term Kv provides a PID gain factor that is proportional to the output velocity. It has the greatest effect on system damping and minimizing oscillations within the drive system. The term usually is the first to be adjusted if you encounter unsatisfactory drive response. Typically, we find Kv to work best in the range of 600 to 800 for lightly to heavily loaded robots, respectively.

The I Term Ki moderates any steady state errors thereby limiting velocity fluctuations during the course of a move. At rest, your robot will seek to “zero out” any command position error. Too large of a Ki factor will cause an excessive windup of the motor when the load changes, such as when climbing over a bump or accelerating to a new speed. Consequently, we typically use a minimum value for Ki in the range of 0 to 10 for lightly to heavily loaded robots respectively.

Encoder and Revcount

P2OS uses the encoder and revcount parameters to convert your platform-independent speed and rotation commands—typically expressed in millimeters or degrees per second, respectively—into platform-dependent units. In previous versions, the encoder value was contained in the P2OS code. With improvements to the motion control servers in P2OS and to support the increasing variety of ActivMedia robot configurations, the encoder value is now included as a configuration parameter, although unlike revcount, its value is a constant.

The encoder value is the number of encoder pulses (“ticks”) per millimeter of wheel rotation. The encoder value is, of course, dependent upon the wheel encoder’s resolution, the motor-to-wheel gear ratio, and the wheel’s diameter.

The revcount value is the number of encoder ticks for one full revolution of the robot. It depends on a number of factors, principally the length of the wheel base, which may change due to payload, tire wear, operating surface, and so on. Accordingly, we provide a calibration tool that lets you determine revcount for your individual robot configuration and operating conditions.

Table 16. Some platform-dependent robot parameter values

Model

52

Page 58
Image 58
Pioneer 2 / PeopleBot manual PID Parameters, Encoder and Revcount

2 / PeopleBot specifications

Pioneer 2, also known as PeopleBot, is an advanced mobile robot platform heralded for its versatile design and robust capabilities. Developed by the renowned robotics company Adept Technology, Pioneer 2 has become a staple in the field of mobile robotics, widely utilized for research, education, and practical applications in various industries.

One of the standout features of Pioneer 2 is its exceptional mobility. The robot is equipped with differential steering, allowing it to navigate complex environments with precision. Its compact and sturdy chassis enables it to traverse a variety of terrains, making it suitable for indoor and outdoor exploration. This mobility is further enhanced by an adaptable wheel configuration, allowing for smooth movement even over obstacles.

In terms of technologies, Pioneer 2 is outfitted with an array of sensors that facilitate autonomous navigation and obstacle avoidance. These sensors include laser range finders, infrared sensors, and bumpers, which work in tandem to map the surrounding environment and detect potential hazards. This capability is critical for applications in areas such as warehouse automation or reconnaissance tasks, where safe navigation is paramount.

Pioneer 2 also supports extensive software frameworks, notably the Robot Operating System (ROS). This compatibility allows researchers and developers to leverage a vast library of tools and algorithms, expediting the process of programming and deploying robotic applications. Additionally, the platform can be integrated with various payloads, such as cameras or grippers, expanding its functionality further.

Another significant characteristic of Pioneer 2 is its user-friendly design. The robot comes with an easy-to-use programming interface that empowers users with varying levels of expertise to engage with the technology. Educational institutions often employ Pioneer 2 in robotics courses to provide students with hands-on experience in programming and operating robotic systems.

Powering Pioneer 2 is a rechargeable battery system that ensures extended operation times. This feature is particularly advantageous for fieldwork applications where connectivity to a power source may not be readily available. Importantly, the robot’s modular design facilitates maintenance and upgrades, allowing users to keep their systems current with advances in technology.

Overall, Pioneer 2 (PeopleBot) exemplifies innovation in mobile robotics with its outstanding mobility, advanced sensing technologies, compatibility with leading software frameworks, and a design focused on ease of use and adaptability. It continues to be a preferred choice for researchers, educators, and professionals in the rapidly evolving landscape of robotics.