Technology Experiment: Navigation Planner

The navigation planner receives commands from the mission planner to perform various tasks using Network Data Delivery Service (NDDS) software, and sends steering commands to the program arbiter using the Task Control Architecture (TCA). Currently, the tasks available are: a straight rows pattern search, a spiral pattern search, and a go-to-waypoint path. The pattern tasks allow a choice of the space between rows and the speed of the robot.

The spiral pattern task enables clockwise or counterclockwise motion and maximum radius selection. The straight row pattern allows a choice of row length and area to cover. The direction to travel (left or right) is defined by a positive or negative area width. The area to cover can also be drawn from a list of vertices defining a polygonal area.

The pattern searches can be interrupted by a go-to-waypoint command. Any number of go-to-waypoint commands can be given before commanding the robot to return to the pattern. When operating in conjunction with morphin and stereo, any morphin commands to avoid obstacles will override the planner's commands, causing the robot to deviate from its pattern or path. When the obstacle is no longer present, the planner's commands will again take precedence, and the robot will resume its pattern or path. When the pattern is completed, or the waypoint is reached, the robot is commanded to stop. The amount of area covered and the total time required are returned to the user through the mission planner and GUI.

Additional tasks to be implemented in the navigation planner include a maneuver planner, which will guide the robot to not only a specific point but also to a specific orientation at that point. Another task is to move the robot so that a target object is within the workspace of a given sensor. Other  pattern types will also be implemented, along with evaluations of the effects of the tasks on power consumption, power generation, and other resources and goals. Improvements to the current tasks will allow the use of an internal map, where obstacles are considered ahead of time to find more optimal paths. The coverage patterns include means of negotiating obstacles that block off part of the area to be covered.

The Antarctic expedition will test these software components. Eventually, completely autonomous commanding of the mission will be tested, with the robot making decisions about where to drive next and when to inspect target objects more closely. Initially, locomotion tests will help determine the best parameters for enabling Nomad to follow the path it is given, based on the terrain and wheel configuration of the robot. These locomotion tests will also provide data for developing the maneuver planner, by providing information on  the range of motion and steering capabilities of Nomad. Also, the power consumption of the robot, for different terrain and different types of motions will be studied for use in the cost analysis by the mission planner.

Studies of how accurately the robot follows the coverage patterns and paths in Antarctica will help in understanding the uncertainties involved with planning. These uncertainties might arise from terrain effects such as rocks and slopes, obstacles, or inaccuracies in following the path. The power generation capability of a robot depends on its location and if solar panels or wind turbines are used. Therefore, the uncertainties in predicting the robot's location over a given path will determine the uncertainties in predicting the power gain from solar or wind power.

Back to Expedition Objectives

Robotic Search for Antarctic Meteorites 1998
All material on this page is property of NASA and Carnegie Mellon University. Any image or text
taken from this site and incorporated into another document without consent violates the Copyright
Law of the United States and the Berne International Copyright Agreement.
Send comments, questions, or suggestions to Dimitrios Apostolopoulos.
This document prepared by Michael Wagner.