Landing : Athabascau University
  • Blogs
  • The Robotics Primer: Chapter 7

The Robotics Primer: Chapter 7


Uncertainty is not much of a problem for simulated robots because their simulated worlds are probably not as unpredictable as our real physical world that real robots must navigate. Since the simulated worlds must have been created by humans, they may have a lot of consistency that does not exist in the real world. Humans would have also had to create the simulated robots, so they could make the robots better prepared for their simulated world based on their knowledge of what the simulated world is like, and they could probably be able to predict what the robot will encounter. Also, the sensors on a simulated robot could be programmed to be 100% accurate, but the sensors on physical robots pick up noise and sometimes have errors, which make it impossible to achieve 100% accuracy.

Better sensors

Accurate sensors allow robots to make better decisions, so it makes sense that sensors are one of the limiting factors in robot intelligence. If a robot had sensors that are 100% accurate and never have any sensory noise or errors, it would be able to make very good and accurate decisions. However, it may not be possible to create sensors that never have errors and are 100% accurate. Also, adding more and better sensors to a robot by itself will not make the robot smarter, because it also needs to be able to process all that sensory data. Sensors may be one limiting factor in robot intelligence, but computational power is also an important limiting factor.

Robot intelligence

Improving the computing power of robots is probably one of the main things that are required for them to become self-aware and highly intelligent. In humans, the main thing that is responsible for our intelligence is our large and powerful brains. We also have great sensors, but even our senses sometimes fail us. We sometimes see things that are not there, and sometimes we cannot hear certain things if we are in a loud environment. Robots do not need to have perfect or very many sensors to become self-aware or highly intelligent, because even our own sensors are not perfect. The important thing when it comes to intelligence is brain power. Since a robot’s brain is a computer, improving the computational power of robots will make them more intelligent.

If robots do become self-aware and highly intelligent, their intelligence could be like ours in some ways, but it would probably be mostly different. Robots can have many kinds of sensors that provide a very wide range of sensory input. There are probably robots that can see ultraviolet light, hear ultrasonic sounds, or sense magnetic fields. This means that robots probably have to process these different types of sensory input in different ways, and this may lead to robots having a different kind of intelligence than humans.