Multimedia  

 

Volume 20 Issue 07 - Publication Date: 1 July 2001
 
RHex: A Simple and Highly Mobile Hexapod Robot
 
U. Saranli University of Michigan, M. Buehler and D.E. Koditschek University of Michigan
 
In this paper, the authors describe the design and control of RHex, a power autonomous, untethered, compliant-legged hexapod robot. RHex has only six actuators—one motor located at each hip—achieving mechanical simplicity that promotes reliable and robust operation in real-world tasks. Empirically stable and highly maneuverable locomotion arises from a very simple clock-driven, open-loop tripod gait. The legs rotate full circle, thereby preventing the common problem of toe stubbing in the protraction (swing) phase. An extensive suite of experimental results documents the robot's significant "intrinsic mobility"—the traversal of rugged, broken, and obstacle-ridden ground without any terrain sensing or actively controlled adaptation. RHex achieves fast and robust forward locomotion traveling at speeds up to one body length per second and traversing height variations well exceeding its body clearance.
 
Multimedia Key
= Video = Data = Code = Image
 
Extension
Media Type
Description
1
Video showing RHex traversing simple outdoor obstacles (October 1999). RHex has complete power and computational autonomy, and receives navigational commands from a remote control unit. It receives no feedback from the environment, except the local feedback at each hip for PD control of motors. The efficacy of the locomotion results from the interaction of the passive robot dynamics and the environment. (1.6MB)
2
Video showing RHex running over a challenging styrofoam obstacle (December 1999). (0.6MB)
3
Video showing a front view of RHex running over the broken surface (December 1999). This obstacle consists of 72 (6 by 12) 7'' square wood blocks, whose heights were chosen from a uniform random distribution between 4" (10.16 cm) and 12" (30.48 cm), corresponding to 1.16 leg lengths height variation. (1.3MB)
4
Video showing a top view of RHex running over the broken surface (December 1999). (1.3MB)
5
Video showing a simulation of the hexapod with a nonzero body initial velocity (June 1999). The simulation environment we developed, SimSect, performs the integration of the hybrid dynamical system arising from the complaint hexapod model. Only contacts between the feet and the ground are allowed, and the stance/flight transitions for the legs are discrete mode transitions, as opposed to a continuous, penetrating ground model. (0.5MB)
6
Video demonstrating the compliance of RHex's legs (December 1999). These legs are made out of "C" shaped 1cm diameter Delrin rods, directly coupled to each hip actuator. They provide compliance mainly in the radial direction, which is approximately 4500 N/m. A leg length of 17.5cm results in a ground clearance of approximately 10.5cm. (0.4MB)
7
Video showing RHex's basic alternating tripod gait (October 1999). This basic gait can achieve fast and stable forward locomotion in the absence of any explicit stabilization effort. Moreover, due to the symmetry of RHex's the same gait can be used to locomote backward and upside-down as well. (1.0MB)
8
Energetics.tar.gz: Speed, power and specific resistance data for all the experiments presented in the paper. The included script PlotEnergetics.m visualizes the data collection in this archive.
9
Turning.tar.gz: Robot position and velocity data for the turning experiments at 5 different speed settings. The included script AnalyzeTurning.m analyzes and displays visual tracking data.
10
Video showing RHex running over a simple styrofoam obstacle. This obstacle was a 1.22 m long strip of 3" (7.62cm) thick Styrofoam board, cut to 15 cm height. This represents 80\% of the robot's leg length and exceeds it's 10.5 cm ground clearance by 4.6 cm, or almost 50\%. Single obstacle experiments were run in this setup with the exception that the robot was powered autonomously (December 1999). Moreover, the data presented in the paper was collected through visual tracking of the robot, which required the experiments to be run in complete darkness. As a result, this example run is not included in the data sets. (0.5MB)
11
Video showing RHex running over a composite obstacle. This second obstacle was built from construction lumber and consisted of a 10 cm high and 63 cm wide base (as viewn in the sagittal plane) on top of which a 8.5 cm high and 3.5 cm wide block was mounted at a distance of 25 cm from the front and a 12.5 cm high and 8.5 cm wide second block was mounted at a distance of 50 cm from the front. Note that the actual composite obstacle experiments were done indoors, over the same obstacle (December 1999). Again, the data presented in the paper was collected through visual tracking of the robot and this example run is not included in the data sets. (0.9MB)
12
SingleObst.tar.gz: Robot position and velocity data for the single obstacle experiments. The included script AnalyzeSingleObstacle.m analyzes and displays visual tracking data.
13
MultiObst.tar.gz: Robot position and velocity data for the composite obstacle experiments. The included script AnalyzeCompositeObstacle.m analyzes and displays visual tracking data.
14
Fractal.tar.gz: Robot position and velocity data for the fractal surface experiments. The included script AnalyzeFractal.m analyzes and displays visual tracking data.
 
Return to Contents