Intelligence without representation
Artificial Intelligence
An Behavior-based Robotics
Developing Natural Language Interfaces: Processing Human Conversations
Developing Natural Language Interfaces: Processing Human Conversations
Object Recognition from Local Scale-Invariant Features
ICCV '99 Proceedings of the International Conference on Computer Vision-Volume 2 - Volume 2
Adaptive execution in complex dynamic worlds
Adaptive execution in complex dynamic worlds
Distinctive Image Features from Scale-Invariant Keypoints
International Journal of Computer Vision
Empirical explorations of the logic theory machine: a case study in heuristic
IRE-AIEE-ACM '57 (Western) Papers presented at the February 26-28, 1957, western joint computer conference: Techniques for reliability
IJCAI'91 Proceedings of the 12th international joint conference on Artificial intelligence - Volume 1
Hi-index | 0.00 |
A classic autonomous robot is an autonomous agent for open, unpredictable environments. Such an agent is inherently autonomous but not independent. Independence implies unpredictability, which is incompatible with agency. The current robot models – behavior based and artificial intelligence – have not been effective at implementing the classic autonomous robot model due to limitations in their definitions. The artificial intelligence model cannot deal with unpredictable environments, and neither model directly includes the concept of agency. Animal training as a model for robotics has the potential to avoid these problems. Animal training has several advantages. It has an inherent model of agency. Goals and behavior are formally separated into human goals and animal behavior. The animal is autonomous, requiring conversation between human and animal, but it is not an independent entity. A robot designed using this model is an articulate machine, programmed as an agent for the user.