RESEARCH METHOD

Research Method Proposal

[Name of the Institute]

ABSTRACT

Although legged locomotion over a moderately rugged terrain can be accomplished by employing simple reactions to the ground contact information, a more effective approach, which allows predictively avoiding obstacles, requires a model of the environment and a control algorithm that takes this model into account when planning footsteps and leg movements. This article addresses the issues of terrain perception and modeling and foothold selection in a walking robot. An integrated system is presented that allows a legged robot to traverse previously unseen, uneven terrain using only onboard perception, provided that a reasonable general path is known. An efficient method for real-time building of a local elevation map from sparse two-dimensional (2D) range measurements of a miniature 2D laser scanner is described. The terrain mapping module supports a foothold selection algorithm, which employs unsupervised learning to create an adaptive decision surface. The robot can learn from realistic simulations; therefore no a priori expert-given rules or parameters are used. The usefulness of our approach is demonstrated in experiments with the six-legged robot Messor. We discuss the lessons learned in field tests and the modifications to our system that turned out to be essential for successful operation under real-world conditions.

CHAPTER 1: INTRODUCTION

Recently, walking robots have been the subject of great interest because of their ability to access unstructured terrain: legged robots have the ability to climb over difficult obstacles without damaging them. It is obvious that high mobility may be extremely useful in the safety, security, and rescue scenarios (Mae, Arai, Inoue, & Miyawaki, 2002; Virk, Gatsoulis, Parack, & Kherada, 2008). Such missions often involve the use of a teleoperation interface; however, the capabilities of a human operator to sense the terrain and to control the robot are limited. Hence, a walking robot control system should have enough autonomy to exploit high mobility in spite of limited feedback from the remote operator (Estremera, Garcia, & González de Santos, 2002).

Legged locomotion in rugged terrain can be achieved by using a clever design of the robot mechanics—by making the legs compliant and by applying simple reflexes to avoid obstacles and to coordinate the legs (Blickhan, Seyfarth, Geyer, Grimmer, Wagner, et al., 2007). Robots applying these principles do not require precise foot placement and require a minimal amount of external perception to traverse a terrain (Saranli, Buehler, & Koditschek, 2001), but without a model of the environment they spend much time recovering from slippages, collisions, and other situations that could be avoided by adapting the footsteps to the terrain.

If we adopt the planning-based approach to rough terrain legged locomotion, the main tasks of the control system are stability maintenance, coordination of legs, generation of foot trajectories, and selection of proper footholds. Recently, there has been much progress in this area, and nowadays there are walking robots with advanced control systems enabling real-time motion planning and execution in rugged terrain (Kolter, Kimz, & Ng, 2009; Plagemann, Mischke, Prentice, Kersting, Roy, et al., 2009; Rusu, Sundaresan, Morisset, Hauser, Agrawal, et ...