Planar Peg-in-hole Insertion Using a Stiffness Controllable Pneumatic Manipulator

نویسندگان

  • Yong Zhu
  • Eric J. Barth
چکیده

This paper presents a method for the impedance control of a pneumatic manipulator for peg-in-hole tasks without using a load cell. The control methodology presented contains a sliding mode force controller and an adaptive pressure summation relationship, which enforces the controllable natural stiffness of the pneumatic actuator. This is accomplished by utilizing two three-way proportional spool valves for each degree-offreedom instead of a four-way valve typically used in fluid power control. Combinations of intrinsic stiffness provided by the compressibility of air and closed-loop stiffness provided by impedance parameters are studied. Experimental results are shown demonstrating that gentle transition from non-contact to contact tasks can be achieved without the use of a load cell by taking advantage of the intrinsically low stiffness of a pneumatic manipulator. Experimental results are also shown demonstrating sensorless (no load cell) force-guided insertion of a planar peg-in-hole task with position uncertainties (hole location not precisely known). 1.0 INTRODUCTION Compliant manipulation, as in peg-in-hole type assembly tasks, requires the manipulator to follow a motion trajectory as well as a force profile, while making contact with a kinematically constrained environment. The potential advantage of pneumatic actuators is that they have natural compliance, which makes them an ideal candidate for causal interaction with an admittance environment. They have the additional advantage of being capable of measuring force using pressure sensors instead of a load cell located at distal ends of the actuator. As properly noted by Pratt et. al. [1] in their work regarding series elastic actuators: “Most robot designers make the mechanical interface between an actuator and its load as stiff as possible. This makes sense in traditional position-controlled systems, because high interface stiffness maximizes bandwidth and, for non-collocated control, reduces instability. However, lower interface stiffness has advantages as well, including greater shock tolerance, lower reflected inertia, more accurate and stable force control, less damage during inadvertent contact, and the potential for energy storage.” The most apparent property of a pneumatic system is the natural compliance allotted by the compressibility of gas. This compliance can be controlled to give a pneumatically actuated system the ability to follow a kinematically constrained trajectory while following a desired contact force profile. Desai et al. [4] presents a minimum impedance control method to minimize interaction forces of a humanoid arm. The method is proved to be effective by using low impedance gains through good modeling and feed-forward compensation. The Whole Arm Manipulator robot is supposed to have intrinsically low impedance, which is actually the nature of pneumatic actuators. Zhu et al. [2] proposes a method for the impedance control of a pneumatic linear actuator for tasks involving contact interaction. The experimental results of the transition from free motion to contact clearly demonstrate that the stiffness of a pneumatic actuator can be controlled by impedance parameters. In Al-Dakkan’s dissertation [3], a single four-way spool valve is decoupled into two three-way valves, which gives the single actuation degree of freedom an additional control degree of freedom for energy saving. In [2], the same strategy of decoupling the dual function of a four-way spool valve into two three-way valves is also applied and utilized by defining a constant pressure summation objective function to relate the two valves. Although the linear actuator could provide good position tracking performance at different pressure summation levels, no explicit relationship between the natural stiffness and impedance stiffness parameter was shown in the contact tasks. The goal of the work presented here is to design a pneumatic actuation system for planar peg-in-hole insertion by taking advantage of a combination of intrinsic (or open-loop) stiffness

برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

منابع مشابه

The Spatial Peg–in–Hole Problem

In this paper the modeling and simulation of a rectangular peg{in{hole insertion carried out by a six axis manipulator is presented. The dynamic model of the robot coupled with the assembly process model is established, where all possible contact points between the peg and the environment are monitored. Every occurring contact point represents a geometrical constraint on the robots dynamic whic...

متن کامل

Complaint Manipulation for Peg-in-Hole: Is Passive Compliance a Key to the Contact Motion?

Peg in Hole has been extensively researched because it represents a contact task of manipulation that requires both position and force control. Most of the approaches assume a manipulator of serially connected rigid links, therefore, they highly depend on excellent resolution of a force/torque sensor that is normally very expensive and quite noisy, a precise model of a dynamic system, and a ver...

متن کامل

Development of force displaying device using pneumatic parallel manipulator and application to palpation motion

Absfracf-The goal of this study is to develop a mechanical system which display elastic characteristic like stiffness on the surface of human body aiming at applying to palpation simulator. Pneumatic parallel manipulator is employed as a driving mechanism, consequently, it brings capability of minute force displaying property owing to the air compressibility. Compliance control system without u...

متن کامل

Planar Parallel 3-rpr Manipulator

This paper describes the design, construction, and control of a planar three degree-of-freedom (dof) inparallel-actuated manipulator at Ohio University. The actuators are three pneumatic cylinders. Using real-time closed-loop feedback control for each actuator length independently, we develop inverse pose and resolved-rate control for this manipulator. The objective of this work is to implement...

متن کامل

Automatic Property Identification via Parameterized Constraints

— During teleoperation, the automatic identification of remote environment properties has the potential of improving performance by providing task-specific feedback to the operator. Similarly, virtual training systems can be calibrated using such an automatic identification procedure. For those properties which can be described by parameterized constraint equations, this paper provides a method...

متن کامل

ذخیره در منابع من


  با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید

برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

عنوان ژورنال:

دوره   شماره 

صفحات  -

تاریخ انتشار 2005