Reconfigurable Exploratory Robotic Vehicles
- Created on Sunday, 01 July 2001
Instrumentation and mobility units could be concatenated as needed.
A family of rugged, modular, reconfigurable, instrumented robotic vehicles has been proposed for use in exploration of the surfaces of Mars and other remote planets. These or similar vehicles could also be useful in such diverse terrestrial applications as exploration of volcanic craters or other hostile terrain, military reconnaissance, inspection of hazardous sites, or searching for victims of earthquakes, landslides, or avalanches. There might even be a market for simplified versions of these vehicles as toys.
The proposed vehicles are denoted generally by the term Axeln, where n is an even number equal to the number of main wheels. The simplest vehicle of this type would be an Axel2 — a two-main-wheel module that would superficially resemble the rear axle plus rear wheels of an automobile (see Figure 1). In addition to the two main wheels, an Axel2 would include a caster wheel (or flat surfboard) attached to the axle by an actuated caster link. The motion (about three quarters of a circle) of the caster link would be used to control the rotation of the axle in order to tilt, to the desired angle, any sensors that might be mounted on the axle. The Axel2 would hence use the same sensors for forward and backward driving. In the vertical position, the caster link is used as an antenna for wireless communication. Two brushes attached to either side of the caster link wipe dust off the solar cells.
In addition to the sensors, the axle of an Axel2 would house computer modules and three motors and associated mechanisms for driving the main wheels and the caster link. Solar cells would be mounted on the outside of the axle except at a mid-length portion, where an assembly containing the caster-link actuator and two module interfaces (one at each end) would be located. Rechargeable batteries would be placed inside the wheel hubs.
One would construct an Axeln (n > 2) as an assembly of multiple Axel2's plus one or more instrument module(s) connected to each other at the module interfaces (see Figure 2). The module interfaces would contain standardized electrical and mechanical connections, including spring-loaded universal joints, about which the modules could comply to adapt to the terrain. Data would be communicated between modules via fast serial links.
An Axeln would amount to a train carrying n/2 — 1 instrument modules. The instrument modules would contain additional computational units that, in addition to processing of instrument readings, could contribute to coordination of train motion. In other words, the "intelligence" of an Axeln, and thus the sophistication of the maneuvers that it could perform, would increase with n.
The symmetrical design of the modules would enable them to operate in any stable orientation, including upside-down; this feature would contribute to robustness of operation in rough terrain, including the ability to recover after falling off a cliff (in the case of Axeln where n < 6). The simple and modular design of the Axels provides better maneuverability using fewer actuators and sensors and hence lower power requirements than traditional rovers. The system features scalable complexity: the motion control algorithms for the Axel2 are very simple; and as the size of the Axel train grows, the complexity of the algorithms increases.
The proposed vehicles would be designed to diagnose themselves
to detect nonfunctional modules. They would be programmed to travel to robotic
service depots, called assembly stations, where nonfunctional modules would
be disconnected and replaced by functional ones.
This work was done by Issa Nesnas of Caltech for NASA's Jet Propulsion Laboratory. For further information, access the Technical Support Package (TSP) free on-line at www.nasatech.com/tsp under the Machinery & Automation category.
This Brief includes a Technical Support Package (TSP).
Reconfigurable Exploratory Robotic Vehicles (reference ) is currently available for download from the TSP library.
Please Login at the top of the page to download.