广度优先搜索(Breadth-first Search)算法描述:
- 用N表示初始结点列表(N待扩展)
- 如果N为空集,则退出并给出失败信号
- n取为N的第一个结点,并在N中删除结点n,放入已访问结点列表
- 如果n为目标结点,则退出并给出成功信号
- 否则,将n的子结点加到N的末尾,并返回2步
分油问题描述:
两个小孩去打油,一人带了一个一斤的空瓶,另一个带了一个七两、一个三两的空瓶。原计划各打一斤油,可是由于所带的钱不够,只好两人合打了一斤油,在回家的路上,二人想平分这一斤油,可是又没有其它工具。试仅用三个瓶子(一斤、七两、三两)精确地分出两个半斤油来。
分油问题的Java实现:
1、对问题建模:问题中的实体对象是空瓶,建立Bottle对象,包含空瓶容量以及当前油量属性
int volume = 0;
int oil = 0;
2、确定分油的可行方式,并确定分油规则
switch (n) {
case 1: // 瓶子1向瓶子2倒油
if (!bottle1.isEmpty() && !bottle2.isFull()) {
n = bottle2.needForFull() <= bottle1.oil ? bottle2
.needForFull() : bottle1.oil;
newState = new StateNode(bottle1.oil - n, bottle2.oil + n,
bottle3.oil, this);
}
break;
case 2: // 瓶子1向瓶子3倒油
if (!bottle1.isEmpty() && !bottle3.isFull()) {
n = bottle3.needForFull() <= bottle1.oil ? bottle3
.needForFull() : bottle1.oil;
newState = new StateNode(bottle1.oil - n, bottle2.oil,
bottle3.oil + n, this);
}
break;
case 3: // 瓶子2向瓶子1倒油
if (!bottle2.isEmpty() && !bottle1.isFull()) {
n = bottle1.needForFull() <= bottle2.oil ? bottle1
.needForFull() : bottle2.oil;
newState = new StateNode(bottle1.oil + n, bottle2.oil - n,
bottle3.oil, this);
}
break;
case 4: // 瓶子2向瓶子3倒油
if (!bottle2.isEmpty() && !bottle3.isFull()) {
n = bottle3.needForFull() <= bottle2.oil ? bottle3
.needForFull() : bottle2.oil;
newState = new StateNode(bottle1.oil, bottle2.oil - n,
bottle3.oil + n, this);
}
break;
case 5: // 瓶子3向瓶子1倒油
if (!bottle3.isEmpty() && !bottle1.isFull()) {
n = bottle1.needForFull() <= bottle3.oil ? bottle1
.needForFull() : bottle3.oil;
newState = new StateNode(bottle1.oil + n, bottle2.oil,
bottle3.oil - n, this);
}
break;
case 6: // 瓶子3向瓶子1倒油
if (!bottle3.isEmpty() && !bottle2.isFull()) {
n = bottle2.needForFull() <= bottle3.oil ? bottle2
.needForFull() : bottle3.oil;
newState = new StateNode(bottle1.oil, bottle2.oil + n,
bottle3.oil - n, this);
}
break;
}
3、在上述的规则,即广度优先搜索的搜索空间中,寻找目标状态。为提高搜索效率,可记住已经搜索过的节点,避免重复搜索,进入无限循环。
ArrayList<StateNode> stateList = new ArrayList<StateNode>(); // 存放所有出现过的状态的数组
if (!tag) { // 如没出现过 加入状态队列和数组
stateQueue.addFirst(newState);
stateList.add(newState);
}
测试结果:
[10 0 0]
[3 7 0]
[3 4 3]
[6 4 0]
[6 1 3]
[9 1 0]
[9 0 1]
[2 7 1]
[2 5 3]
[5 5 0]
完整代码见附件