All language subtitles for 极狐 S HI 版 长视频 V2_讯飞听见_原始
Afrikaans
Akan
Albanian
Amharic
Arabic
Armenian
Azerbaijani
Basque
Belarusian
Bemba
Bengali
Bihari
Bosnian
Breton
Bulgarian
Cambodian
Catalan
Cebuano
Cherokee
Chichewa
Chinese (Simplified)
Chinese (Traditional)
Corsican
Croatian
Czech
Danish
Dutch
Esperanto
Estonian
Ewe
Faroese
Filipino
Finnish
French
Frisian
Ga
Galician
Georgian
German
Greek
Guarani
Gujarati
Haitian Creole
Hausa
Hawaiian
Hebrew
Hindi
Hmong
Hungarian
Icelandic
Igbo
Indonesian
Interlingua
Irish
Italian
Japanese
Javanese
Kannada
Kazakh
Kinyarwanda
Kirundi
Kongo
Korean
Krio (Sierra Leone)
Kurdish
Kurdish (Soranî)
Kyrgyz
Laothian
Latin
Latvian
Lingala
Lithuanian
Lozi
Luganda
Luo
Luxembourgish
Macedonian
Malagasy
Malay
Malayalam
Maltese
Maori
Marathi
Mauritian Creole
Moldavian
Mongolian
Myanmar (Burmese)
Montenegrin
Nepali
Nigerian Pidgin
Northern Sotho
Norwegian
Norwegian (Nynorsk)
Occitan
Oriya
Oromo
Pashto
Persian
Polish
Portuguese (Brazil)
Portuguese (Portugal)
Punjabi
Quechua
Romanian
Romansh
Runyakitara
Russian
Samoan
Scots Gaelic
Serbian
Serbo-Croatian
Sesotho
Setswana
Seychellois Creole
Shona
Sindhi
Sinhalese
Slovak
Slovenian
Somali
Spanish
Spanish (Latin American)
Sundanese
Swahili
Swedish
Tajik
Tamil
Tatar
Telugu
Thai
Tigrinya
Tonga
Tshiluba
Tumbuka
Turkish
Turkmen
Twi
Uighur
Ukrainian
Urdu
Uzbek
Vietnamese
Welsh
Wolof
Xhosa
Yiddish
Yoruba
Zulu
Would you like to inspect the original subtitles? These are the user uploaded subtitles that are being translated:
1
00:00:00,040 --> 00:00:03,410
5 月 7 号极狐阿尔法S全新 HI 版车型正式上市了
2
00:00:03,730 --> 00:00:06,140
时隔一年我们终于等到了这一刻
3
00:00:06,150 --> 00:00:08,100
我相信你们肯定也等了很久
4
00:00:08,420 --> 00:00:10,410
很幸运的是我们这次收到了邀请
5
00:00:10,420 --> 00:00:11,570
去北京体验了
6
00:00:11,570 --> 00:00:14,970
最新状态的阿尔法 S 和这套智能驾驶
7
00:00:15,160 --> 00:00:15,990
不过很遗憾的是
8
00:00:16,000 --> 00:00:18,270
上海还在疫情居家的状态下
9
00:00:18,280 --> 00:00:20,790
所以我也只能在家来录这期视频
10
00:00:20,800 --> 00:00:22,710
大家看我头发可能也能看出
11
00:00:22,720 --> 00:00:24,740
我应该很久没有出门了
12
00:00:24,750 --> 00:00:26,300
如果实音的效果没有特别好
13
00:00:26,510 --> 00:00:27,940
也希望大家能够多多见谅
14
00:00:28,070 --> 00:00:29,900
这一次我们北京的
15
00:00:29,910 --> 00:00:31,300
辅助驾驶测试员小伙伴
16
00:00:31,430 --> 00:00:32,740
参加了这次活动
17
00:00:33,010 --> 00:00:35,150
所以华为的 ADS 能力上
18
00:00:35,160 --> 00:00:36,020
到底有没有变化
19
00:00:36,030 --> 00:00:38,340
我们又体验到了什么新的功能
20
00:00:38,580 --> 00:00:41,190
过去这一年极狐跟华为都干了啥
21
00:00:53,530 --> 00:00:56,200
这次我们在北京体验了这套辅助驾驶
22
00:00:56,210 --> 00:00:57,280
也是华为 ADS
23
00:00:57,290 --> 00:01:00,480
第一次在上海以外的城市公开展示
24
00:01:00,700 --> 00:01:01,970
虽然是在北京体验
25
00:01:01,980 --> 00:01:04,490
但是五环内是没有高精地图数据的
26
00:01:04,500 --> 00:01:06,530
所以无法使用导航辅助驾驶
27
00:01:06,540 --> 00:01:09,050
所以我们需要自驾到五环高架入口
28
00:01:09,060 --> 00:01:10,450
再开启这个功能
29
00:01:11,050 --> 00:01:12,150
然后我们可以看到
30
00:01:12,160 --> 00:01:13,890
这个标志已经亮起来了
31
00:01:13,900 --> 00:01:17,820
我们可以双击智驾导航辅助已激活
32
00:01:17,830 --> 00:01:18,980
智驾导航辅助已激活
33
00:01:21,580 --> 00:01:23,130
它现在就是自动提示变道
34
00:01:23,380 --> 00:01:25,250
然后我们可以轻转方向盘确认换道
35
00:01:25,520 --> 00:01:26,560
但因为侧后向有车
36
00:01:26,570 --> 00:01:28,320
所以说它会适当减速
37
00:01:28,490 --> 00:01:29,640
找机会去换
38
00:01:29,650 --> 00:01:32,550
后面又来了一辆车
39
00:01:33,610 --> 00:01:35,720
一直在动态的调整速度来找
40
00:01:36,010 --> 00:01:37,120
最合适的机会跟进去
41
00:01:37,120 --> 00:01:38,400
找到机会了
42
00:01:39,080 --> 00:01:39,820
好变过来了
43
00:01:41,720 --> 00:01:43,510
环路的路况对于这套系统
44
00:01:43,520 --> 00:01:45,110
自然没有太大的挑战
45
00:01:45,120 --> 00:01:47,750
不过测试的时候刚好开始下雨了
46
00:01:47,880 --> 00:01:50,150
挂在激光雷达和摄像头前的水滴
47
00:01:50,160 --> 00:01:51,150
和扬起的水雾
48
00:01:51,240 --> 00:01:53,310
会对传感器带来比较大的干扰
49
00:01:53,320 --> 00:01:56,790
这也促成了第一次 ADS 的雨天测试
50
00:01:57,360 --> 00:01:57,830
这会雨不大
51
00:01:58,000 --> 00:02:00,710
但是因为刚刚上午下了比较大的雨
52
00:02:00,720 --> 00:02:02,030
所以地上有很多积水
53
00:02:02,240 --> 00:02:05,950
把周围的车辆它溅起来的这些水雾
54
00:02:05,960 --> 00:02:07,150
还是比较重的
55
00:02:07,540 --> 00:02:09,100
这些水雾其实对于激光雷达来说
56
00:02:09,110 --> 00:02:10,100
影响是很大的
57
00:02:10,450 --> 00:02:13,290
所以说还是需要用比较重的 AI 的算法
58
00:02:13,300 --> 00:02:14,250
来去除掉这些雨雾
59
00:02:16,660 --> 00:02:17,670
然后我们现在自动驾驶
60
00:02:17,670 --> 00:02:19,070
雨突然变大了
61
00:02:19,420 --> 00:02:20,830
我们这时候要自动使出来了
62
00:02:22,900 --> 00:02:24,850
不过无论是环路还是市区
63
00:02:24,860 --> 00:02:27,010
当前的雨量对这套系统的工作
64
00:02:27,100 --> 00:02:28,730
都没有任何的影响
65
00:02:29,270 --> 00:02:30,860
下环路之后是一个左转
66
00:02:30,860 --> 00:02:33,300
系统起步速度很快
67
00:02:33,970 --> 00:02:35,180
即将完成转弯的时候
68
00:02:35,470 --> 00:02:37,980
后方有辆车想汇入我方车道
69
00:02:37,980 --> 00:02:39,660
系统迅速做出了避让
70
00:02:42,160 --> 00:02:43,390
来到一个新路口
71
00:02:43,400 --> 00:02:44,950
面对右侧合流的车辆
72
00:02:44,950 --> 00:02:46,670
系统预测了行驶轨迹
73
00:02:46,680 --> 00:02:48,630
提前做出了避让的动作
74
00:02:48,740 --> 00:02:50,970
这波操作确实很有安全感
75
00:02:50,980 --> 00:02:54,090
接下来又是两个左转和两个右转
76
00:02:54,090 --> 00:02:56,130
系统的操作行云如流水
77
00:02:56,140 --> 00:02:58,530
很遗憾的是北京因为疫情的原因
78
00:02:58,540 --> 00:03:01,290
很多人都进入了居家办公的状态
79
00:03:01,300 --> 00:03:02,650
所以路上比较通畅
80
00:03:02,660 --> 00:03:05,970
对于华为 ADS 来说没有太大的挑战
81
00:03:06,250 --> 00:03:07,720
大家的疑问可能和我一样
82
00:03:07,730 --> 00:03:09,160
这一年到底进步了吗
83
00:03:09,170 --> 00:03:10,600
华为都干了啥
84
00:03:11,120 --> 00:03:13,310
首先在现场可以体验到的是
85
00:03:13,320 --> 00:03:14,550
这套系统的风格
86
00:03:14,640 --> 00:03:15,990
已经更加娴熟
87
00:03:16,000 --> 00:03:17,430
更像一个老司机了
88
00:03:17,440 --> 00:03:19,710
整个体验过程中没有突兀的点刹
89
00:03:19,720 --> 00:03:20,630
每一次决策
90
00:03:20,840 --> 00:03:23,070
也都跟人类司机非常相似
91
00:03:23,080 --> 00:03:25,380
准备在这进行一个左转
92
00:03:26,270 --> 00:03:27,220
左转路口很大
93
00:03:28,820 --> 00:03:30,530
它检到黄灯了
94
00:03:30,540 --> 00:03:32,490
所以它就决定了不闯红灯了
95
00:03:32,860 --> 00:03:33,970
不去抢黄灯了
96
00:03:33,980 --> 00:03:35,170
多变的城市道路中
97
00:03:35,300 --> 00:03:36,810
系统不再刻板的保持
98
00:03:36,820 --> 00:03:38,410
在车道中间行驶
99
00:03:38,630 --> 00:03:40,180
如果右侧近距离有车
100
00:03:40,190 --> 00:03:42,660
系统会主动偏向左侧行驶
101
00:03:42,930 --> 00:03:45,160
这种让驾驶员非常有安全感的
102
00:03:45,160 --> 00:03:48,040
拟人化小细节贯穿了整个行程
103
00:03:49,480 --> 00:03:51,030
咱们这大车你看我们的方向盘
104
00:03:51,040 --> 00:03:52,030
都会微微的调整
105
00:03:52,040 --> 00:03:53,070
非常轻微的调整
106
00:03:53,380 --> 00:03:55,150
来和这些车保持一个安全距离
107
00:03:55,950 --> 00:03:57,420
前面过来了一个老大爷
108
00:03:57,990 --> 00:03:59,330
看我们车能不能检出来
109
00:03:59,330 --> 00:04:00,370
检出老大爷
110
00:04:00,380 --> 00:04:02,610
并且我们的轨迹线做了一个避让
111
00:04:02,970 --> 00:04:03,800
我们再往下走
112
00:04:03,810 --> 00:04:04,960
避让这个老大爷
113
00:04:05,990 --> 00:04:07,380
我们目前体验过的
114
00:04:07,380 --> 00:04:09,700
点到点的智能驾驶的车型来说
115
00:04:09,710 --> 00:04:12,900
华为 ADS 这个是目前体验最好的
116
00:04:13,230 --> 00:04:14,660
我们要在这汇出
117
00:04:14,670 --> 00:04:15,940
你看它很智能的选择了
118
00:04:15,950 --> 00:04:17,580
左边这个车道汇出
119
00:04:17,590 --> 00:04:19,180
是因为它知道接下来
120
00:04:19,190 --> 00:04:21,060
马上我们还要靠左走
121
00:04:21,410 --> 00:04:23,080
所以它没有选择去右边那条车道
122
00:04:23,090 --> 00:04:23,920
然后再换回来
123
00:04:24,160 --> 00:04:25,430
另一点就是泛化
124
00:04:25,440 --> 00:04:26,030
在此之前
125
00:04:26,120 --> 00:04:29,350
这套系统的主要测试跟调校都在上海
126
00:04:29,360 --> 00:04:30,430
但是每个地区
127
00:04:30,520 --> 00:04:33,510
都有每个地区的道路风格和驾驶风格
128
00:04:33,610 --> 00:04:35,760
为了让这套系统在每一个城市
129
00:04:35,770 --> 00:04:36,960
都有比较好的体验
130
00:04:36,970 --> 00:04:39,600
工程师需要不断的测试优化
131
00:04:39,820 --> 00:04:40,970
截止到发布会之前
132
00:04:41,060 --> 00:04:42,010
华为已经完成了
133
00:04:42,020 --> 00:04:44,370
120 万公里的泛化测试
134
00:04:44,380 --> 00:04:46,250
如果到这里体验就结束了
135
00:04:46,260 --> 00:04:47,850
我相信你们肯定不接受
136
00:04:47,850 --> 00:04:48,610
这什么玩意儿
137
00:04:48,620 --> 00:04:50,970
还不如把去年 4 月的视频再看一遍
138
00:04:51,570 --> 00:04:52,520
看看进度条
139
00:04:52,520 --> 00:04:53,120
这才哪到哪
140
00:04:54,250 --> 00:04:55,440
这次华为给我们带来了一个
141
00:04:55,440 --> 00:04:58,040
全新的功能 AVP 代客泊车
142
00:04:58,530 --> 00:04:59,640
这个功能有一点
143
00:04:59,650 --> 00:05:01,880
类似于小鹏的停车场记忆泊车
144
00:05:01,890 --> 00:05:03,880
但是两者最大的区别是
145
00:05:03,890 --> 00:05:05,760
小鹏记录的是一条路线
146
00:05:05,770 --> 00:05:08,760
而华为记忆的是整个停车场的模型
147
00:05:08,880 --> 00:05:10,590
所以实现的功能和能力
148
00:05:10,720 --> 00:05:12,670
都是一个更高的维度
149
00:05:12,970 --> 00:05:15,600
首先这个功能在任何一个地下停车场
150
00:05:15,610 --> 00:05:16,600
都是可以使用的
151
00:05:16,610 --> 00:05:18,440
它不需要任何场端设备
152
00:05:18,450 --> 00:05:21,280
完全依赖车身自身的传感器实现
153
00:05:21,420 --> 00:05:22,810
所以在使用之前
154
00:05:22,940 --> 00:05:24,850
需要手动行驶进入地库
155
00:05:24,860 --> 00:05:26,250
把停车场扫描一次
156
00:05:26,250 --> 00:05:27,770
才可以使用 AVP
157
00:05:27,780 --> 00:05:30,210
说白了就是带它认个路
158
00:05:30,550 --> 00:05:31,420
认路的时候
159
00:05:31,430 --> 00:05:33,100
也不需要手动开启这个功能
160
00:05:33,110 --> 00:05:35,180
你只需要按照你日常驾驶的习惯
161
00:05:35,190 --> 00:05:37,500
正常驾驶停到你想要去的车位
162
00:05:37,510 --> 00:05:40,700
系统就会在后台默默记录下这个停车场
163
00:05:41,670 --> 00:05:43,600
它提示发现了新的收藏点
164
00:05:43,820 --> 00:05:45,790
收藏后可用于自动泊车
165
00:05:46,270 --> 00:05:48,870
我们尝试收藏一下
166
00:05:49,940 --> 00:05:51,490
这一点真的是非常赞
167
00:05:51,500 --> 00:05:52,610
自动化程度很高
168
00:05:52,620 --> 00:05:53,730
体验也很无感
169
00:05:54,070 --> 00:05:56,100
之前基础自动泊车的评测中
170
00:05:56,230 --> 00:05:57,500
我们就已经提到过了
171
00:05:57,510 --> 00:05:58,940
自动化程度越高
172
00:05:58,950 --> 00:06:00,420
使用门槛越低
173
00:06:00,430 --> 00:06:01,660
只有门槛越低
174
00:06:01,670 --> 00:06:03,300
功能才有价值
175
00:06:03,640 --> 00:06:04,470
进入地库之后
176
00:06:04,640 --> 00:06:06,710
系统可以通过摄像头和雷达
177
00:06:06,710 --> 00:06:08,350
实时扫描整个停车场
178
00:06:08,350 --> 00:06:11,710
这套系统同样可以实现跨楼层的能力
179
00:06:12,010 --> 00:06:13,280
把整个停车场扫完之后
180
00:06:13,410 --> 00:06:15,280
系统会用自身的计算芯片
181
00:06:15,290 --> 00:06:17,400
建出整个停车场的模型
182
00:06:17,620 --> 00:06:18,410
存储在本地
183
00:06:18,410 --> 00:06:20,010
算的速度也非常快
184
00:06:20,020 --> 00:06:21,690
基本没有等待的时间
185
00:06:21,690 --> 00:06:23,010
可以看到这边
186
00:06:23,020 --> 00:06:24,010
AVP 地图构建中
187
00:06:24,010 --> 00:06:25,030
速度很快
188
00:06:25,040 --> 00:06:25,830
已经 90% 了
189
00:06:26,520 --> 00:06:29,770
说话功夫已经 95%
190
00:06:30,310 --> 00:06:34,190
下面是地图可用于 AVP 自动代客泊车
191
00:06:38,370 --> 00:06:39,840
这张图已经构建好了
192
00:06:40,000 --> 00:06:41,310
第二次来到这个停车场
193
00:06:41,320 --> 00:06:43,110
则可以直接使用 AVP 功能
194
00:06:43,120 --> 00:06:46,550
让系统自动从入口开到收藏的车位
195
00:06:46,560 --> 00:06:48,990
自动建出了我们刚才构好的这张图
196
00:06:49,000 --> 00:06:49,790
这是二层的图
197
00:06:49,800 --> 00:06:50,910
一会可能就切换三层
198
00:06:50,920 --> 00:06:54,510
好我们试一下自动泊车已激活
199
00:06:54,760 --> 00:06:55,990
然后它开始走了
200
00:06:56,410 --> 00:06:57,560
刚才我们也提到了
201
00:06:57,570 --> 00:07:00,160
华为记录的是整个停车场的模型
202
00:07:00,170 --> 00:07:02,640
相比记录一条路线的小鹏来说
203
00:07:02,730 --> 00:07:04,160
则有两大优势
204
00:07:04,350 --> 00:07:05,940
第一个就是在建图的时候
205
00:07:06,110 --> 00:07:07,820
没有行驶距离的限制
206
00:07:07,830 --> 00:07:10,060
而是以建图的面积来衡量
207
00:07:10,170 --> 00:07:11,080
它不只是按照距离
208
00:07:11,080 --> 00:07:12,120
它是整个面积
209
00:07:12,130 --> 00:07:13,600
整个是一个大面积几层面积
210
00:07:13,610 --> 00:07:14,520
全部把它建起来
211
00:07:15,040 --> 00:07:17,230
中间的道路坡道全部都构建出来
212
00:07:17,240 --> 00:07:18,830
不过具体面积多大
213
00:07:18,840 --> 00:07:21,630
我们目前得到的答案是上千平
214
00:07:21,640 --> 00:07:23,710
另外行驶的路线也没有了
215
00:07:23,720 --> 00:07:25,350
不要交叉的需求
216
00:07:25,650 --> 00:07:28,240
第二点就是有了整个停车场的模型
217
00:07:28,250 --> 00:07:30,120
在地库入口你可以直接选择
218
00:07:30,130 --> 00:07:31,480
你想要停的车位
219
00:07:31,610 --> 00:07:34,080
目标车位不局限在收藏的车位
220
00:07:34,090 --> 00:07:35,480
和沿途的车位
221
00:07:35,490 --> 00:07:37,360
如果你选择的是一个新的车位
222
00:07:37,370 --> 00:07:38,640
去往目标车位的路线
223
00:07:38,770 --> 00:07:40,160
和之前记忆的路线
224
00:07:40,250 --> 00:07:41,480
也没有任何关系
225
00:07:41,490 --> 00:07:44,830
系统会非常聪明的选择一条最优的路线
226
00:07:45,340 --> 00:07:47,090
自动泊车已激活
227
00:07:47,220 --> 00:07:48,610
它这次要自动抄近道了
228
00:07:48,620 --> 00:07:49,810
它就不上那边绕了
229
00:07:50,180 --> 00:07:51,250
又抄了个近道
230
00:07:51,260 --> 00:07:52,730
而且最惊艳的是
231
00:07:52,740 --> 00:07:54,530
如果你选择的车位有车
232
00:07:54,530 --> 00:07:57,720
系统会主动停到周围有空的车位
233
00:07:57,730 --> 00:07:58,680
这点简直是 YYDS
234
00:07:58,680 --> 00:08:01,620
太聪明了
235
00:08:01,620 --> 00:08:02,820
目标车位被占
236
00:08:03,250 --> 00:08:04,510
自动寻找空车位
237
00:08:05,040 --> 00:08:06,310
它给我们找了另外一个空车位
238
00:08:06,320 --> 00:08:07,990
看到这里可能有人会问
239
00:08:08,000 --> 00:08:09,670
如果每个停车场在用之前
240
00:08:09,760 --> 00:08:10,990
都要完整的扫一次
241
00:08:11,240 --> 00:08:12,390
岂不是很蛋疼
242
00:08:12,620 --> 00:08:14,690
这里就是华为体验做得好的地方了
243
00:08:14,700 --> 00:08:17,410
首先不需要扫描整个完整的停车场
244
00:08:17,420 --> 00:08:19,290
完全可以扫多少用多少
245
00:08:19,300 --> 00:08:20,890
其次工作人员也表示
246
00:08:21,330 --> 00:08:23,240
接下来他们会在热门地段
247
00:08:23,250 --> 00:08:24,800
商超去扫一些建图
248
00:08:24,810 --> 00:08:26,560
这样用户即使没有去过
249
00:08:26,570 --> 00:08:28,880
到了之后也是可以直接使用的
250
00:08:29,160 --> 00:08:30,270
但是比较遗憾的就是
251
00:08:30,280 --> 00:08:32,030
这次停车场比较空旷
252
00:08:32,040 --> 00:08:33,590
所以我们并没有体验到
253
00:08:33,600 --> 00:08:35,110
避让绕行的能力
254
00:08:35,250 --> 00:08:37,240
这一点只能在后期体验了之后
255
00:08:37,450 --> 00:08:38,520
再来补充
256
00:08:38,530 --> 00:08:39,600
完整的一镜到底
257
00:08:39,810 --> 00:08:40,880
我们也已经发布了
258
00:08:40,880 --> 00:08:41,760
感兴趣的朋友
259
00:08:41,850 --> 00:08:42,920
可以去看一看
260
00:08:43,430 --> 00:08:44,700
不过就泊车能力来说
261
00:08:44,790 --> 00:08:46,260
我们上次已经体验到了
262
00:08:46,260 --> 00:08:49,740
泊入速度基本上稳在 3 km/h 非常快
263
00:08:49,750 --> 00:08:51,340
另外即使你找了一个
264
00:08:51,350 --> 00:08:52,660
非常刁钻的车位
265
00:08:52,660 --> 00:08:55,460
系统仍然能够规划出路线停进去
266
00:08:55,610 --> 00:08:58,560
这一点真的是远超了很多人类司机
267
00:08:58,770 --> 00:09:01,280
而且系统还会把行驶路径的意图
268
00:09:01,370 --> 00:09:03,200
展示在可视化的界面上
269
00:09:03,210 --> 00:09:05,120
这一点要给绝对的好评
270
00:09:05,530 --> 00:09:06,960
就是这个全景的界面
271
00:09:06,960 --> 00:09:08,840
真的说不上是好看
272
00:09:09,910 --> 00:09:12,980
到这里所有的体验就已经全部结束了
273
00:09:12,990 --> 00:09:13,940
不出意外的是
274
00:09:13,940 --> 00:09:15,260
华为再次出场
275
00:09:15,260 --> 00:09:16,700
又刷新了我们的认知
276
00:09:16,710 --> 00:09:18,900
而且随着 AVP 功能推出
277
00:09:18,910 --> 00:09:20,510
华为也无限接近了
278
00:09:20,510 --> 00:09:23,230
停车场到停车场的点到点智能驾驶
279
00:09:23,650 --> 00:09:25,440
所以现在压力就来到了
280
00:09:25,450 --> 00:09:27,960
小鹏 理想 蔚来这些新势力车企上
281
00:09:27,970 --> 00:09:29,320
尤其是理想跟蔚来
282
00:09:29,330 --> 00:09:31,040
还没有展示过这些能力
283
00:09:31,240 --> 00:09:33,150
虽然小鹏是第一个提出
284
00:09:33,150 --> 00:09:35,390
停车场到停车场智能驾驶概念的
285
00:09:35,400 --> 00:09:36,670
但就目前而言
286
00:09:36,680 --> 00:09:38,510
华为 BETA 版的系统体验
287
00:09:38,520 --> 00:09:40,110
确实会更好一些
288
00:09:40,590 --> 00:09:42,020
城市导航辅助驾驶的市场
289
00:09:42,150 --> 00:09:44,060
仍然处于起步的初期
290
00:09:44,070 --> 00:09:46,940
泛化的难度也远超我们的预期
291
00:09:47,110 --> 00:09:48,340
比尔盖茨曾经说过
292
00:09:48,350 --> 00:09:50,140
我们总是高估一年
293
00:09:50,150 --> 00:09:51,860
或者两年中能够做到的
294
00:09:51,870 --> 00:09:54,620
而低估五年或者十年中能够做到的
295
00:09:54,820 --> 00:09:57,690
所以我们不应该抱着短期一步到位
296
00:09:57,700 --> 00:10:00,090
全自动驾驶的想法看待辅助驾驶
297
00:10:00,100 --> 00:10:02,250
但也不应该低估辅助驾驶
298
00:10:02,260 --> 00:10:03,730
长期对我们的价值
299
00:10:03,980 --> 00:10:06,770
OK 以上就是本期视频的全部内容了
300
00:10:06,780 --> 00:10:08,170
我们也非常迫切的希望
301
00:10:08,180 --> 00:10:09,530
极狐能够早日的交付
302
00:10:09,540 --> 00:10:11,010
因为高阶辅助驾驶能力
303
00:10:11,140 --> 00:10:12,970
有太多无法传达的感受
304
00:10:13,120 --> 00:10:15,150
一次体验足以颠覆认知
305
00:10:15,160 --> 00:10:16,070
这里是 42 号车库
306
00:10:16,070 --> 00:10:16,990
我是 63
307
00:10:17,000 --> 00:10:18,670
如果你觉得这期视频还不错的话
308
00:10:18,920 --> 00:10:20,470
一定要给我们一个一键三连
309
00:10:20,480 --> 00:10:21,790
我们下期再见
310
00:10:21,800 --> 00:10:23,270
也希望上海早日解封
311
00:10:23,270 --> 00:10:23,430
拜拜
21563
