| M. C. Torrance. The case for a realistic mobile robot simulator. In Working Notes of the AAAI Fall Symposium on Applications of Artificial Intelligence to Real-World Autonomous Mobile Robots, Cambridge, MA, October, 1992. |
....simulations over physical experiments is that they are repeatable, faster (simulating a 1 hour experiment takes about 5 minutes) and do not suffer unexpected hardware breakdown. The disadvantages in terms of model faithfulness are, of course, well known (for a more complete discussion of this see [57]) In order to determine the importance of the imitative following strategy for the success of the learning, simulations were carried out with two types of learner agents, some with and some without the ability of following. The agents had to learn three different types of vocabularies: 1) a ....
M. C. Torrance. The case for a realistic mobile robot simulator. In Working Notes of the AAAI Fall Symposium on Applications of Artificial Intelligence to Real-World Autonomous Mobile Robots, Cambridge, MA, October, 1992.
....over physical experiments was that they were repeatable, faster (simulating a 1 hour experiment takes about 5 minutes) and did not suffer unexpected hardware breakdown. The disadvantages in terms of model faithfulness are, of course, well known (for a more complete discussion of this see [41]) In our experiments, physical and simulated worlds differ in many aspects. For instance in the simulation a poor account is given of the physics of the sensors and of the world perceived by the robot (a simple field of view is defined for the light dispersion whose intensity is invariant over ....
Torrance M. C. (1992), `The Case for a Realistic Mobile Robot Simulator," in Working Notes of the AAAI Fall Symposium on Applications of Artificial Intelligence to Real-World Autonomous Mobile Robots, Cambridge, MA, October.
....over physical experiments was that they were repeatable, faster (simulating a 1 hour experiment takes about 5 minutes) and did not suffer unexpected hardware breakdowns. The disadvantages in terms of model faithfulness are, of course, well known (for a more complete discussion of this see Torrance, 1992). In our experiments, physical and simulated worlds differ in many aspects. For instance in the simulation a poor account is given of the physics of the sensors and of the world perceived by the robot (a simple field of view is defined for the light dispersion whose intensity is invariant over ....
Torrance, M.C.. (1992). The case for a realistic mobile robot simulator. Working Notes of the AAAI Fall Symposium on Applications of Artificial Intelligence to RealWorld Autonomous Mobile Robots. Cambridge, MA.
Online articles have much greater impact More about CiteSeer.IST Add search form to your site Submit documents Feedback
CiteSeer.IST - Copyright Penn State and NEC