ROS-robot-programming-book-by-turtlebo3-developers-EN(三)
The high level of English is a standard for a top student.
1. How do you improve your sleeping quality?
Well, there are several ways. First of all, a good night sleep or a sound sleep has everything to do with the exercise you do during the day. So physical exercise is vital. Secondly, you can always take some medication. If you suffer from a really bad insomnia, you can take some sleeping pills. Though, personally, I am afraid of the side-effects. But I take them anyway because it is the only way I can sleep through the night. Thirdly, a cup of warm milk is always helpful. Those are the ways we can use to improve our sleeping quality.
2. There are songs that can really make you sad and cry when you hear them. but it’s actually not the song that makes you cry, it’s the people behind the memories.
3. gain new knowledge by reviewing old.
4. Obstacles don’t have to stop you. If you run into a wall, don’t turn around and give up. Figure out how to climb it, go through it, or work around it.
5. You quietly into my heart, but in the end, kept distance with me.
6. Today's advantage will be replaced by tomorrow's trend, grasp the trend, grasp the future.
7. Too many people think of happiness as the ultimate goal of life. But, if you're waiting for happiness to arrive then it's likely that it never will!
8. Like sunlight, sunset, we appear, we disappear. We are so important to some, but we are just passing through.
9. Live without pretending, love without depending, listen without defending, speak without offending.
10. A life with love will have some thorns, but a life without love will have no roses.
11. I have no regrets in my life. I think that everything happens to you for a reason. The hard times that you go through build character, making you a much stronger person.
12. All lives end, all hearts are broken. Caring is not an advantage!
13. We should all start to live before we get too old.
Reading books （317~409 page）
Two simulation methods of the TurtleBot3 package were introduced. One is to use RViz, a 3D visualization tool of ROS, and the other is to use Gazebo, a 3D robot simulator. Simulation is a great tool for users because it allows performing programming tasks very close to the actual environment with a robot.
What do we need to implement navigation in robots? It may vary depending on the navigation algorithm, and the followings may be required as basic features.
➊ Map
➋ Pose of Robot
➌ Sensing
➍ Path Calculation and Driving
SLAM(Simultaneous Localization And Mapping) is developed to let the robot create a map with or without a help of a human being. This is a method of creating a map while the robot explores the unknown space and detects its surroundings and estimates its current location as well as creating a map.
Paper
1. The Closed-Loop Control System Design of Hexapod Robot Autonomous Navigation Based on Fuzzy Neural Network
(http://robot.sia.cn/CN/10.13973/j.cnki.robot.170252)
Abstract: For the autonomous navigation issue of hexapod robot in unknown environment, an autonomous navigation closed-loop control algorithm based on fuzzy neural network is proposed, and a navigation control system of hexapod robot is designed according to the algorithm. The algorithm combines the logical reasoning ability of fuzzy control with the learning and training ability of neural network, and introduces the closed-loop control method into optimization. The control system is composed of four modules:information input, fuzzy neural network, instruction execution and information feedback. The perception of environment and location information is completed by GPS (global positioning system) sensor, electronic compass sensor and ultrasonic sensor. The fuzzy neural network control algorithm is reconstructed by C language and is applied to the system. The simulation results show that the closed-loop control algorithm based on fuzzy neural network is better than the open-loop control algorithm theoretically. The closed-loop control algorithm can reduce the walking distance of the hexapod in the environment with obstacles, and the walking speed is increased by 6.14%, the walking time is shortened by 8.74%. On this basis, a physical test is carried out. The experimental results show that the control system can achieve the autonomous obstacle-avoidance control of the hexapod robot. Compared with the open-loop control system, the walking path can be shortened effectively, the walking speed is increased by 5.66% and the walking time is shortened by 7.25%, which verifies the feasibility and practicability of the closed-loop control system.
Key words： hexapod robot autonomous navigation fuzzy neural network closed-loop control environmental perception
2. Deep Learning for Control: The State of the Art and Prospects
Abstract: Deep learning has shown great potential and advantage in feature extraction and model fitting. It is significant to use deep learning for control problems involving high dimension data. Currently, there have been some investigations focusing on deep learning in control. This paper is a review of related work including control object recognition, state feature extraction, system parameter identification and control strategy calculation. Besides, this paper describes the approaches and ideas of deep control, adaptive dynamic programming and parallel control related to deep learning in control. Also, this paper summarizes the main functions and existing problems of deep learning in control presents some prospects of future work. Key words Deep learning, control, feature, adaptive dynamic programming (ADP)
