behavior planning——15.cost function design weightTweaking
Designing cost functions is difficult and getting them all to cooperate to produce reasionable vehicle behavior is hard.
some of the difficulties associated with cost function design include solving new problems without unsolving old ones.
when you are working on a self-driving car,you may find that the vehicle is behaving reasonably well except for some
particular situations.Maybe it is not being aggressive enough about making left turns at traffic lights.
So, in an effort to solve this problem, you either add new cost funcions, tweak existing ones, or modify weights.But every time you do, there is chance that you will introduce some breaking change into something that already works.
in pracice ,we solve this throuth regression testing, where we define some set of situations, Then ,whenever we make a change, we simulate the viheicle in all of our test cases and make sure that it still behaves expected.
We won't say more about testing here, but it is an important part of developing software in a safety-cricical application.
The next difficulty is balancing costs of drastically different magnitudes.
Because, yes ,we want to get to our destination efficiently, but if we are in a situation where safety is an issue,
we want to solve that problem and not think about efficiency at all. one way to do that is to have weights which reflect
the type of problem the cost function addresses. so we want to most heavily penalize any behavior which simply is not possible due to physics, then we want to think about safety ,legality ,comfort.
And only once those are satisfied do we want to think about efficiency. But we alse may want to adjust the relative importance if these weights depending on situation. If a light turns red ,for example,legality becomes a much more relevant
concern than when we engage in normal highway driving . this leads us to out last difficulty, reasoning about individual cost functions. Ideally, each cost function will serve a very specific responsibility.
which is something we do not do in our earlier example of s speed cost function.
We were trying to balance out desire to drive quickly, which has to do with efficiency, with out desire to not exceed the speed limit , which is legality. In practice, we might want to difine several cost functions associated with vehicle speed.
In which case we might have a binary cost function which just checks to see if we are breaking the speed limit and the continuous cost function which pulls us towards out target speed.
By assigning each cost function to a very specific role, like safety versus legality versus efficiency, we can then standardize the output of all cost functions to be between -1 and 1. Additionlly, It is helpful to paraetrize whenever possible.
This allows us to use some parameter optimization technique like gradient descent along with our set of regression tests to
programmatically tweak our cost functions. Finally ,thinking in terms of vehicle state is helpful. The things we can indirestly
control about our vehicle are it is position, velocity, and acceleration. It can be helpful to keep these in mind when coming up with cost functions. Let is walk through an example.
Say we want to think about the following classer of cost functions.
And to make it easier, to keep everything straight,let is think in terms of position, velocity ,and acceleration.
The binary "Are we exceeding the speed? "cost functin would go here, then the cost function that wanted to keep us close
to the speed limit would go here. and now ,instead of being that weird discontinuous gragh we made earlier, this could just be some parabola like this.
And even though the cost of this is low even for speeds that exceed the speed limit, that is OK because we have our
binary cost function which would prevent that behavior. Continuing to think about speed, we also might want to try to
drive at a speed that is close to the average speed of traffic, for safety reasons ,even if that speed is above or below the
speed limit. And that would go here.
In the position column, we would have an obvious feasibility concern which is collision avoidance.
We can not drive somewhere if there is already a car there. Then, for safety reasons, wo would want the buffer distance,
which tries to keep us far from other vehicles, and the cost function which checks to make sure we are driving on the road
near the center of our lane and in a lane that is close to our goal lane.
For acceleration, we would first want to make sure we only consider behaviors that the car can execute,
Make sure we only consider behaviors that the car can execute, and then we would want to avoid having any rapid changes in acceleration because those are perceived as uncomfortable.
This is also known as jerk as which you will learn more about in the next lesson.
consider a merge onto a highway, for example.
This is a potentially dangerous situation where we really want to get up to traffic speed as quickly as possible.
So this cost function may become more relevant than it mormally is. But we alse want to make sure that we yield if there
is not really a gap, so we want to make sure this cost funcion and this one are weighted sufficiently high.
and we can compare these merge priorities to a different situation.
for examle, a car approaching a green light that suddenly turns yellow.
In this situation, we would probably want to boost the weights associated with legality and would probably need to
add a whole new cost function for obeying traffic rules.
now that do not fall neatly into the position, velocity ,or acceleration classes, so I will just put it over here.
If this is all starting to feel like it is getting pretty complex ,well,you are right.
It is pretty hard to aviod this exploding complexity when using finite state machines.
Partially that is because of the finite state machine itself, but we are also trying to solve a very hard problem and
some complexity is unavoidable no matter what solution approach you take.
behavior planning——15.cost function design weightTweaking的更多相关文章
- behavior planning——13. implement a cost function in C++
In the previous quizzes, you designed a cost function to choose a lane when trying to reach a goal i ...
- behavior planning——11 create a cost function speed penalty
A key part of getting transitions to happen when we want them to is the design of reasonable cost ...
- behavior planning——14.implement a cost function in C++
n most situations, a single cost function will not be sufficient to produce complex vehicle behavior ...
- behavior planning——12.example cost funtion -lane change penalty
In the image above, the blue self driving car (bottom left) is trying to get to the goal (gold sta ...
- Model Representation and Cost Function
Model Representation To establish notation for future use, we’ll use x(i) to denote the “input” vari ...
- 吴恩达机器学习笔记 - cost function and gradient descent
一.简介 cost fuction是用来判断机器预算值和实际值得误差,一般来说训练机器学习的目的就是希望将这个cost function减到最小.本文会介绍如何找到这个最小值. 二.线性回归的cost ...
- loss function与cost function
实际上,代价函数(cost function)和损失函数(loss function 亦称为 error function)是同义的.它们都是事先定义一个假设函数(hypothesis),通过训练集由 ...
- 【caffe】loss function、cost function和error
@tags: caffe 机器学习 在机器学习(暂时限定有监督学习)中,常见的算法大都可以划分为两个部分来理解它 一个是它的Hypothesis function,也就是你用一个函数f,来拟合任意一个 ...
- 逻辑回归损失函数(cost function)
逻辑回归模型预估的是样本属于某个分类的概率,其损失函数(Cost Function)可以像线型回归那样,以均方差来表示:也可以用对数.概率等方法.损失函数本质上是衡量”模型预估值“到“实际值”的距离, ...
随机推荐
- 集训队日常训练20180525-DIV2
A.2295 求有多少素数对和等于n. 暴力. #include <bits/stdc++.h> using namespace std; int ss(int n) { ,a=sqrt( ...
- sprite学习
CSS雪碧图,就是把所有的图表,按钮和图形包含在一个图像里面.它要求: 静态图片,不随用户信息变化而变化:小图片,图片容量比较小:加载量比较大. 使用这种技术可以减少Web浏览器发出的服务器请求,显著 ...
- 数据库通过sql备份脚本恢复时,报错误The user specified as a definer ('root'@'%') does not exist
数据库通过sql备份脚本恢复时,报错误The user specified as a definer ('root'@'%') does not exist 当出现这个错误,意思是某个数据库对象的定义 ...
- Mysql查询优化-DB篇
本文重点从数据库本身角度,硬件和环境的优化不在本文范围内 1. 使用索引(Index All Columns Used in 'where', 'order by', and 'group by' C ...
- Android原生调用mui里面的js如何实现
遍历所有运行中的webview页面,采用自带的SDK方法进行获取所有的IWebview.MUI中自带的webview是一个IWebviewArrayList<IWebview> webli ...
- Vue调试神器之Vue.js devTools
Vue项目中使用Vue.js devTools这款调试神器,可以极大程度的提高我们的开发效率. 安装 1.打开Chrome网上应用商店安装插件(自墙),直接搜索devTools安装即可.贵宾传送阵,请 ...
- day38 21-今天的内容总结
以前我们在web层里面去调Service再在Service里面去调DAO是一路new过去的,在web层里面new Service,然后在Service里面new DAO.每次你都需要主动去找这个对象. ...
- 微服务开源生态报告 No.6
「微服务开源生态报告」,汇集各个开源项目近期的社区动态,帮助开发者们更高效的了解到各开源项目的最新进展. 社区动态包括,但不限于:版本发布.人员动态.项目动态和规划.培训和活动. 非常欢迎国内其他微服 ...
- bash 小技巧
CTRL-R (reverse find),按下之后敲几个字母就能在最近打过的命令里搜索.
- oralce如何修改默认的XDB监听端口
Oracle9i默认的XML DB把HTTP的默认端口设为8080,这是一个太常用的端口了,很多别的WebServer都会使用这个端口, 如果我们安装了它,最好修改一下,避免冲突,如果不使用呢,就最好 ...