Research - Requirements
Adaptron must be able to learn and think using any combination of the following robot features.
The ultimate objective is to handle all the higher complexity settings.
| Features | Low Complexity | High Complexity |
|---|---|---|
| Number of senses | One | Multiple |
| Sensors per sense | One | Multiple |
| Multiple Sensors | Independent sensors | Dependent sensor array - one dimension |
| Sensor array | Linear | Circular |
| Number of sensor arrays | One Dimension | Two Dimensions (not currently a requirement) |
| Type of readings / inputs | Symbolic (Nominal) | Magnitude (Ratio scale ) |
| Number of action device types | One | Multiple |
| Number of action devices per type | One | Multiple |
| Simultaneous actions | One at a time per device | Multiple in parallel |
|
Value Systems (Motivation –> Behaviour) |
Familiar / Unfamiliar (Intrinsic –> Exploration) |
(Extrinsic –> Survival) |
| Thinking ahead | One stimulus | Multiple steps |
| Time flow | Event driven | Clock driven |
Other constraints that have been imposed on the design of Adaptron are:
- It is to be programmed using a deterministic algorithm, no probabilities
- It is to be as simple as possible; Occam’s Razor applies
- It is to start with no experience; it must learn everything; 100% grounded, embodied and enactive
- It is to operate within a robot body which is separate from the environment
- The robot body shall house the sensors and action devices
- It is to start with two built in value systems, one for exploration and the other for survival
- It shall use reinforcement learning driven by the value systems for learning actions
- It will use unsupervised learning for perception ( not supervised learning )