Chapter one first introduced the definition of applying AI technology to robotics. An intelligent robot should be capable of learning and adapting to unstructured environments and changing its behavior based on the situations even though the actions are not included in the program. This reminds me of the robots we have done in class. Even though we wrote many codes for the robots when encountering different cases, they did not have the ability to self-learning and alter their behaviors automatically. The author also mentioned The Turing Test, a test for distinguishing if the robot can listen and understand the order, and answer the question in a human way. Moreover, the author introduced the “OODA” loop, which stands for Observe-Orient-Decide-Act. In this loop, the robot needs to collect data using sensors, put the data in a reference frame, analyze and make decisions, and put it into action. In this case, the users can set goals for the robot, and the robots can gradually achieve them, just like the picking-up toys example mentioned in the book. By the way, in the description of making a picking-up toys robot, there is one piece of advice I found may be useful to solve the previous problems I and Vivian met. “If you have noise from the motors upsetting the Arduino (and you will tell because the Arduino will keep resetting itself), you can add a small filter capacitor of 10 μ across the motor wires “(Govers 17). In the process of working on Cherokey 4WD, we faced a similar problem. We can try this possible solution when next time.
Chapter ten summarised the idea of subsumption architecture, which divided the robot’s goals into different levels. The lower level means simpler and more fundamental tasks, while the upper level’s goal is more difficult and advanced. The concept reminds me of the functional diagram I just finished for my final project. I can not tell which diagram is better but I think they can help the designs to think of the order to produce robots. In addition, he indicates the possibilities of robotics in the future and explained that nowadays only the hardware nor the software is advancing. Even though many believe robots will substitute human work or even control humans, the author believes the issue is just an exaggeration. The author believes that because robots have no demand based on Maslow’s Hierarchy of Needs, they really do not have the motivation to evolve. Furthermore, he emphasizes the importance of robotics to avoid and prevent risks because humans think robots should not make mistakes. I agreed with him that robotics brings more opportunities than threats. In the article “Robot Smog ” we read before, the author put forward the idea that it becomes easier for people to make robots, which may bring about chaos. However, after half semester of study, I feel like it is really hard to build a robot, not to mention an intelligent robot. Therefore, in the foreseen future, I guess the work opportunities would increase due to the developing technology of robotics, while at the same time robots do not have the ability or demand to evolve ito a level that humans can not stop them.
Leave a Reply