Trajectory Panning and Control of Robot Arm for Planetary Surface Sample Missions

Paper number

IAC-09.A3.I.2

Author

Dr. Chakravarthini M. Saaj, Surrey Space Centre, University of Surrey, United Kingdom

Coauthor

Mr. Soheil Seyyedi Parsa, United Kingdom

Coauthor

Mr. Abbas Khorram, United Kingdom

Year

2009

Abstract

The conventional motion planning and control algorithms for mobile robots are not suitable for autonomous planetary rovers moving on rough terrains. The exploration of Mars for traces of past or present life and presence of water and minerals in the subsurface rocks remains one of the most challenging and exciting areas for the scientific community. The use of fully autonomous robotic missions on Mars is vital for unmanned exploration of Mars and will be likely precursors to future manned missions. Mars sample return missions has taken a further step toward realisation with the recent publication of a report by space agencies across the world on an International Mars Architecture for the Return of Samples (iMARS). 

A dextrous planetary manipulator could be considered as an important part of any surface rover payload for surface science experiment and sample acquisition. It is an essential part of any in-situ analysis or sample return mission. The minimum requirement is for a three degree of freedom teleoperated arm with a scoop for trench digging, but the addition of wrist pitch with a gripper gives much greater flexibility of deployment. Science experiments usually include analytical probing such as point spectrometry, instrument placement (like Alpha Particle X-Ray Spectrometer, Nuclear magnetic resonance or Mossbauer spectrometer), material sampling such as trench digging, core sampling and sample acquisition and containment. 

Having an absolute control over the planetary manipulator for executing the assigned task is a vital requirement of any planetary mission design. Controlling the manipulator as a system requires a precise knowledge of the system behaviour over the time and this could be achieved by having accurate kinematic and dynamic models. Accurate kinematical and dynamical model of serial link manipulator drives the manipulator’s joint actuators to accomplish a desired task. 

A six degree of freedom, Mitsubishi MELFA RV-1A robot arm is used through out this study to validate the trajectory planning and control algorithms. RV-1A’s mathematical model based on forward and inverse kinematics of the manipulator has been developed and would be used with a class of polynomial functions to generate a sequence of time-based control set points for the trajectory planning. The joint controller is designed based on the dynamic model to achieve the desired system performance for sample collection in real-time. It is envisaged that the proposed modelling technique, trajectory planning method and control algorithm would prove useful for sample acquisition and transfer for landers, rovers and sample return missions. A detailed theoretical analysis together with experimental results will be presented in the final paper.

Abstract document

IAC-09.A3.I.2.pdf

Manuscript document

IAC-09.A3.I.2.pdf (🔒 authorized access only).

To get the manuscript, please contact IAF Secretariat.