当前位置:网站首页>Ind wks first week
Ind wks first week
2022-07-03 10:31:00 【IFI_ rccsim】
Inspired by other projects in charge , Think Dubins The curve is feasible in the player's specific path planning .
1、Dubins Curve overview
Dubins The curve is in the case of accessibility , Provide an optimal forward path , And A* ,RRT Compare with those search algorithms , Another advantage is that the planned path meets the kinematic requirements .
( The picture below is taken from the Internet ) yes Dubins Six basic forms of curves , The first four are most commonly used , In a word, a generalization is a vector with a known starting point ( Coordinates and directions ) And the vector of the target point , With the help of dubins Curve algorithm can solve the shortest path .

2、 Why do you think of transplantation Dubins curve ?
(1)Dubins Curves were first used in vehicle motion planning , It is also widely used in the field of unmanned aerial vehicles , According to my current basic understanding of simulation players , These are similar , Include : The forward direction is generally forward and the U-turn action is complex 、 There is a limit to the maximum turning curvature when turning 、 The motion state of a point at a certain time can be abstracted as a vector .
(2)Dubins The curve is significantly better than the common A* Equal basic algorithm , And it is basically derived by mathematical calculus , Low dependence on search , Furthermore, the algorithm complexity is low .
(3)robocup The player's penalty mechanism is to decide the penalty according to the number of collisions when there is a collision , Therefore, if we can minimize the collision that does not happen because of grabbing the ball , You can reduce the number of times you are sent off .
(4) Watch the code of other teams and find , There are often automatic falls during walking , Part of the situation is related to the excessive turning angle , If you use this curve , Then the maximum curvature can be given by the student who designed the action after the experiment , Can reduce the situation of self falling .
(5) This curve itself is relatively simple , The basic type is “ arc —— A straight line —— arc ”, It does not involve actions such as reversing and turning around in place, which have a large loss of speed , And then make the displacement as large as possible in a short time .
3、 Thinking about whether transplantation is meaningful
(1) By observing the team code game , At present, it is difficult to see a similar curve trajectory in the team path .
(2) As the ball moves , The target position and direction of players change very fast , And this curve track changes greatly when the target point , It may be difficult to cross to a straight line when traveling all the way along the arc segment .
(3) in the light of (2), Have you thought about it onball Applied ideas , I also thought of the idea of experimenting in the allocation of players far from the ball .
4、 Difficulties encountered
(1) Not familiar with the team code , Failed to call existing functions skillfully , Often look for code for a long time .
(2) Concerns about effectiveness , If it takes more time to complete the experiment, but the effect is not good , It may slow down the progress of the whole team .
边栏推荐
- QT creator uses OpenCV Pro add
- Implementation of "quick start electronic" window dragging
- 20220604数学:x的平方根
- Hands on deep learning pytorch version exercise answer - 2.2 preliminary knowledge / data preprocessing
- Data preprocessing - Data Mining 1
- 【毕业季】图匮于丰,防俭于逸;治不忘乱,安不忘危。
- 20220601数学:阶乘后的零
- Notes - regular expressions
- Raspberry pie 4B deploys lnmp+tor and builds a website on dark web
- Julia1.0
猜你喜欢

Deep learning by Pytorch

Ut2015 learning notes

重写波士顿房价预测任务(使用飞桨paddlepaddle)

A super cool background permission management system

Leetcode - 5 longest palindrome substring

Ut2013 learning notes

一个30岁的测试员无比挣扎的故事,连躺平都是奢望

LeetCode - 900. RLE iterator

Powshell's set location: unable to find a solution to the problem of accepting actual parameters

Policy Gradient Methods of Deep Reinforcement Learning (Part Two)
随机推荐
Hands on deep learning pytorch version exercise solution - 2.6 probability
Leetcode刷题---977
20220604数学:x的平方根
Leetcode-100: same tree
Discrete-event system
Inverse code of string (Jilin University postgraduate entrance examination question)
Simple real-time gesture recognition based on OpenCV (including code)
Seata分布式事务失效,不生效(事务不回滚)的常见场景
[C question set] of Ⅵ
Powshell's set location: unable to find a solution to the problem of accepting actual parameters
20220609其他:多数元素
Ut2015 learning notes
20220608其他:逆波兰表达式求值
【SQL】一篇带你掌握SQL数据库的查询与修改相关操作
MySQL报错“Expression #1 of SELECT list is not in GROUP BY clause and contains nonaggre”解决方法
神经网络入门之模型选择(PyTorch)
What useful materials have I learned from when installing QT
Advantageous distinctive domain adaptation reading notes (detailed)
【毕业季】图匮于丰,防俭于逸;治不忘乱,安不忘危。
Leetcode-112:路径总和