加拿大丛林迷路五小时，ChatGPT救命神技，比地图还靠谱

新智元07-15

新智元报道编辑：英智【新智元导读】想象一下，你在加拿大的森林里迷路5小时，手机电量只剩3%，Google Maps失灵，信号微弱。但ChatGPT靠实时坐标救了场，堪称AI导航的教科书，快来围观。最近，X平台上一个帖子火了，一群人在加拿大偏远的小镇Mabou骑全地形车（ATV）玩，结果迷路了整整五小时，最后靠ChatGPT导航才安全回来。Mabou是个小地方，周围全是没开发过的森林和小路。他们本来...

网页链接

免责声明：本文观点仅代表作者个人观点，不构成本平台的投资建议，本平台不对文章信息准确性、完整性和及时性做出任何保证，亦不对因使用或信赖文章信息引发的任何损失承担责任。

精彩评论

我们需要你的真知灼见来填补这片空白

发表看法

{"i18n":{"language":"zh_CN"},"isChannel":false,"data":{"share":"https://www.laohu8.com/m/news/2551190803?lang=zh_CN&edition=full","thumbnail":"","is_english":false,"pubTime":"2025-07-15 21:03","share_image_url":"https://static.laohu8.com/e9f99090a1c2ed51c021029395664489","id":"2551190803","market":"us","top_or_hot":-1,"title":"加拿大丛林迷路五小时，ChatGPT救命神技，比地图还靠谱","media":"新智元","content":"<div>\n<p>新智元报道编辑：英智【新智元导读】想象一下，你在加拿大的森林里迷路5小时，手机电量只剩3%，Google Maps失灵，信号微弱。但ChatGPT靠实时坐标救了场，堪称AI导航的教科书，快来围观。最近，X平台上一个帖子火了，一群人在加拿大偏远的小镇Mabou骑全地形车（ATV）玩，结果迷路了整整五小时，最后靠ChatGPT导航才安全回来。Mabou是个小地方，周围全是没开发过的森林和小路。他们本来...</p>\n\n<a href=\"https://tech.ifeng.com/c/8l17HvR25Vt\">网页链接</a>\n\n</div>\n","source":"fenghuang_stock","html":"<!DOCTYPE html>\n<html>\n<head>\n<meta http-equiv=\"Content-Type\" content=\"text/html; charset=utf-8\" />\n<meta name=\"viewport\" content=\"width=device-width,initial-scale=1.0,minimum-scale=1.0,maximum-scale=1.0,user-scalable=no\"/>\n<meta name=\"format-detection\" content=\"telephone=no,email=no,address=no\" />\n<title>加拿大丛林迷路五小时，ChatGPT救命神技，比地图还靠谱</title>\n<style type=\"text/css\">\na,abbr,acronym,address,applet,article,aside,audio,b,big,blockquote,body,canvas,caption,center,cite,code,dd,del,details,dfn,div,dl,dt,\nem,embed,fieldset,figcaption,figure,footer,form,h1,h2,h3,h4,h5,h6,header,hgroup,html,i,iframe,img,ins,kbd,label,legend,li,mark,menu,nav,\nobject,ol,output,p,pre,q,ruby,s,samp,section,small,span,strike,strong,sub,summary,sup,table,tbody,td,tfoot,th,thead,time,tr,tt,u,ul,var,video{ font:inherit;margin:0;padding:0;vertical-align:baseline;border:0 }\nbody{ font-size:16px; line-height:1.5; color:#999; background:transparent; }\n.wrapper{ overflow:hidden;word-break:break-all;padding:10px; }\nh1,h2{ font-weight:normal; line-height:1.35; margin-bottom:.6em; }\nh3,h4,h5,h6{ line-height:1.35; margin-bottom:1em; }\nh1{ font-size:24px; }\nh2{ font-size:20px; }\nh3{ font-size:18px; }\nh4{ font-size:16px; }\nh5{ font-size:14px; }\nh6{ font-size:12px; }\np,ul,ol,blockquote,dl,table{ margin:1.2em 0; }\nul,ol{ margin-left:2em; }\nul{ list-style:disc; }\nol{ list-style:decimal; }\nli,li p{ margin:10px 0;}\nimg{ max-width:100%;display:block;margin:0 auto 1em; }\nblockquote{ color:#B5B2B1; border-left:3px solid #aaa; padding:1em; }\nstrong,b{font-weight:bold;}\nem,i{font-style:italic;}\ntable{ width:100%;border-collapse:collapse;border-spacing:1px;margin:1em 0;font-size:.9em; }\nth,td{ padding:5px;text-align:left;border:1px solid #aaa; }\nth{ font-weight:bold;background:#5d5d5d; }\n.symbol-link{font-weight:bold;}\n/* header{ border-bottom:1px solid #494756; } */\n.title{ margin:0 0 8px;line-height:1.3;color:#ddd; }\n.meta {color:#5e5c6d;font-size:13px;margin:0 0 .5em; }\na{text-decoration:none; color:#2a4b87;}\n.meta .head { display: inline-block; overflow: hidden}\n.head .h-thumb { width: 30px; height: 30px; margin: 0; padding: 0; border-radius: 50%; float: left;}\n.head .h-content { margin: 0; padding: 0 0 0 9px; float: left;}\n.head .h-name {font-size: 13px; color: #eee; margin: 0;}\n.head .h-time {font-size: 11px; color: #7E829C; margin: 0;line-height: 11px;}\n.small {font-size: 12.5px; display: inline-block; transform: scale(0.9); -webkit-transform: scale(0.9); transform-origin: left; -webkit-transform-origin: left;}\n.smaller {font-size: 12.5px; display: inline-block; transform: scale(0.8); -webkit-transform: scale(0.8); transform-origin: left; -webkit-transform-origin: left;}\n.bt-text {font-size: 12px;margin: 1.5em 0 0 0}\n.bt-text p {margin: 0}\n</style>\n</head>\n<body>\n<div class=\"wrapper\">\n<header>\n<h2 class=\"title\">\n加拿大丛林迷路五小时，ChatGPT救命神技，比地图还靠谱\n</h2>\n\n<h4 class=\"meta\">\n\n\n2025-07-15 21:03 北京时间&nbsp;&nbsp;&nbsp;<a href=https://tech.ifeng.com/c/8l17HvR25Vt><strong>新智元</strong></a>\n\n\n</h4>\n\n</header>\n<article>\n<div>\n<p>新智元报道编辑：英智【新智元导读】想象一下，你在加拿大的森林里迷路5小时，手机电量只剩3%，Google Maps失灵，信号微弱。但ChatGPT靠实时坐标救了场，堪称AI导航的教科书，快来围观。最近，X平台上一个帖子火了，一群人在加拿大偏远的小镇Mabou骑全地形车（ATV）玩，结果迷路了整整五小时，最后靠ChatGPT导航才安全回来。Mabou是个小地方，周围全是没开发过的森林和小路。他们本来...</p>\n\n<a href=\"https://tech.ifeng.com/c/8l17HvR25Vt\">网页链接</a>\n\n</div>\n\n\n</article>\n</div>\n</body>\n</html>\n","isBrief":false,"type":0,"news_type":1,"symbol":"LU0320765059.SGD","symbol_name":"FTIF - Franklin US Opportunities A Acc SGD","start_time":0,"source_url":"https://tech.ifeng.com/c/8l17HvR25Vt","article_id":"2551190803","we_media_id":null,"thumbnails":[],"rights":null,"url":"https://stock-news.laohu8.com/highlight/detail?id=2551190803","pubTimestamp":1752584636,"columns":[],"sourceInfo":{"source_id":"fenghuang_stock","name":"凤凰网"},"weMediaInfo":null,"summary":"最近，X平台上一个帖子火了，一群人在加拿大偏远的小镇Mabou骑全地形车玩，结果迷路了整整五小时，最后靠ChatGPT导航才安全回来。随时调整路线他们每5-10分钟发一次新坐标，ChatGPT就根据最新位置调整建议。下图显示，ChatGPT确认他们还在正轨上，建议继续沿着Mabou河走，还给出了下一步选择。有人夸ChatGPT“救命神器”，但也有人吐槽：有手机信号和GPS，干嘛不直接看卫星图？ChatGPT不光会聊天，还能靠实时数据给实用建议，传统地图失灵时特别管用。","collect":0,"end_time":0,"defaultTopTitle":"ifeng.com","property":[],"viewcount":null,"language":"zh","relate_stocks":{"LU0320765059.SGD":"FTIF - Franklin US Opportunities A Acc SGD","ATV":"橡果国际","LU1267930573.SGD":"TEMPLETON GLOBAL \"AA\" (SGD) ACC A","LU1815336760.USD":"THREADNEEDLE (LUX) GLOBAL TECHNOLOGY \"AUP\" (USD) INC","LU2023250843.SGD":"Allianz Thematica Cl AT Acc H2-SGD","LU0528227936.USD":"富达环球人口趋势基金A-ACC","LU2420271590.USD":"ALLIANZ SELECT INCOME AND GROWTH \"AT\" (USD) ACC","LU0323591593.USD":"SCHRODER ISF QEP GLOBAL QUALITY \"A\" (USD) ACC","LU0820561909.HKD":"ALLIANZ INCOME AND GROWTH \"AM\" (HKD) INC","LU1121112475.USD":"NINETY ONE GSF GLOBAL FRANCHISE \"A\" (USDHDG) ACC","SG9999014914.USD":"UNITED GLOBAL QUALITY GROWTH (USDHDG) INC","LU1267930730.SGD":"富兰克林美国机遇基金AS Acc SGD (CPF)","LU0316494557.USD":"FRANKLIN GLOBAL FUNDAMENTAL STRATEGIES \"A\" ACC","SG9999015952.SGD":"LIONGLOBAL DISRUPTIVE INNOVATION \"I\" (SGD) ACC","LU1868837136.USD":"CT (LUX) I AMERICAN \"8\" (USD) ACC","LU1119994496.HKD":"FIDELITY WORLD \"A\" (HKD) ACC","LU1506573853.SGD":"MANULIFE GF GLOBAL EQUITY \"AA\" (SGD) INC","LU0094547139.USD":"abrdn SICAV I - GLOBAL SUSTAINABLE EQUITY FUND \"A\" (USD) ACC","LU1435385759.SGD":"Natixis Loomis Sayles US Growth Equity RA SGD-H","LU0312595415.SGD":"Schroder ISF Global Climate Change Equity A Acc SGD","LU0882574139.USD":"富达环球消费行业基金A ACC","LU0080751232.USD":"富达环球多元动力基金A","LU0640476718.USD":"THREADNEEDLE (LUX) US CONTRARIAN CORE EQ \"AU\" (USD) ACC","BK4576":"AR","LU0029864427.USD":"TEMPLETON GLOBAL \"A\" (USD) INC","IE00BQXX3D17.EUR":"GUINNESS GLOBAL INNOVATORS \"C\" (EUR) ACC","LU2168564065.EUR":"AZ ALLOCATION - TREND \"AAZ\" (EUR) ACC","LU2362541513.USD":"WELLINGTON NEXT GENERATION GLOBAL EQUITY \"A\" (USD) ACC","LU0345769128.USD":"NINETY ONE GSF GLOBAL EQUITY \"A\" (USD) ACC","LU0943347566.SGD":"安联收益及增长平衡基金AM H2-SGD","LU1935042215.USD":"MANULIFE GF GLOBAL MULTI-ASSET DIVERSIFIED INCOME  \"AA\" (USD) INC A","LU1261432733.SGD":"Fidelity World A-ACC-SGD","LU1880398471.USD":"AMUNDI FUNDS GLOBAL EQUITY \"A2\" (USD) ACC","LU2404859667.USD":"FUNDSMITH EQUITY FUND \"R\" (USD) ACC","LU0820562030.AUD":"ALLIANZ INCOME AND GROWTH \"AMH2\" (AUDHDG) H2 INC","LU0477156953.USD":"HARRIS ASSOCIATES US VALUE EQUITY \"RE\" (USD) ACC","LU1244550577.SGD":"FTIF - Franklin Global Multi-Asset Income A (Mdis) SGD-H1","BK4612":"AI芯片","LU1988902786.USD":"FULLERTON LUX FUNDS GLOBAL ABSOLUTE ALPHA \"I\" (USD) ACC","LU0792757196.USD":"TEMPLETON SHARIAH GLOBAL EQUITY FUND \"A\" (USD) ACC","LU2063271972.USD":"富兰克林创新领域基金","LU0345768153.USD":"NINETY ONE GSF GLOBAL STRATEGIC MANAGED \"A\" (USD) ACC","LU0310800965.SGD":"FTIF - Templeton Global Balanced A Acc SGD","LU0444973449.USD":"CT (LUX) I GLOBAL TECHNOLOGY \"DU\" (USD) ACC","LU1551013425.SGD":"Allianz Income and Growth Cl AMg2 DIS H2-SGD","LU0353189680.USD":"富国美国全盘成长基金Cl A Acc","LU1496350171.SGD":"FRANKLIN DIVERSIFIED BALANCED \"A\" (SGDHDG) ACC","LU0994945656.USD":"NINETY ONE GSF GLOBAL FRANCHISE \"A\" (USD) INC 2","LU1496350502.SGD":"FRANKLIN DIVERSIFIED DYNAMIC \"A\" (SGDHDG) ACC"},"translate_title":"Lost in the Canadian jungle for five hours, ChatGPT's life-saving skills are more reliable than maps","themeId":null,"isJumpTheme":false,"ttsUrl":null,"symbols_score_info":{"ATV":1},"content_text":"新智元报道编辑：英智【新智元导读】想象一下，你在加拿大的森林里迷路5小时，手机电量只剩3%，Google Maps失灵，信号微弱。但ChatGPT靠实时坐标救了场，堪称AI导航的教科书，快来围观。最近，X平台上一个帖子火了，一群人在加拿大偏远的小镇Mabou骑全地形车（ATV）玩，结果迷路了整整五小时，最后靠ChatGPT导航才安全回来。Mabou是个小地方，周围全是没开发过的森林和小路。他们本来打算从Upper Southwest Mabou Rd出发，骑18公里到Whycocomagh玩一圈。计划挺好，可他们不小心偏离了主路，钻进了地图上没标的小径。结果Google Maps、ATV专用app啥的都用不了，因为这些工具只认主干道，压根没收录那些隐秘小路，害他们彻底迷路了。手机信号也不好，电量只剩3%，其中一个骑手突发奇想，尝试用ChatGPT来导航，每5-10分钟给ChatGPT发一次GPS坐标。没想到这招还真救了他们。ChatGPT一步步救援第一张截图中，ChatGPT给出了清晰又接地气的回复，把路线拆成了几个简单步骤。发坐标求助他们用手机GPS把实时位置（45.9697°N，61.4119°W）发给ChatGPT，说想去Whycocomagh（45.96435°N，61.1426°W），问有没有能走的路。ChatGPT给力回应ChatGPT分析了坐标和地形，给了超清晰的步步指引：第一步：从上西南马布路往东走，这条路是条泥土路，ATV能过。第二步：汇入Chestico Trail / Celtic Shores Coastal Trail（切斯蒂科步道/凯尔特海岸步道），这步道从东边开始，合法给ATV用，约17公里，从Fort Hood到Mabou河。第三步：沿着步道往东走，平行19号公路，穿过林子，跟着Mabou河走。第四步：快到Mabou村时，往南或东拐，接上Route 252（通往Whycocomagh的路），就到啦！这些指引不仅有路名，还告诉你方向（东、南）和地形（林子、河），超实用，迷路的人一看就懂。随时调整路线他们每5-10分钟发一次新坐标，ChatGPT就根据最新位置调整建议。下图显示，ChatGPT确认他们还在正轨上，建议继续沿着Mabou河走，还给出了下一步选择。碾压传统导航Google Maps和ATV app因为没小径数据彻底扑街。而ChatGPT靠GPS坐标加卫星视图，猜出了小路怎么连起来，再用大白话给出指引。这简直是救命稻草，特别适合野外。有人夸ChatGPT“救命神器”，但也有人吐槽：有手机信号和GPS，干嘛不直接看卫星图？帖子的回复解释了，卫星图虽然能看，但没法像ChatGPT这样实时给个性化文字指引，尤其在复杂地形里。研究证明，像ChatGPT这样的LLM在户外导航上真有潜力。比如PathGPT框架，把历史路线变成文字，再用AI生成个性化路线的办法，效果杠杠的。PathGPT：向朋友问路想象一下，打开导航APP说：“找一条从公司到家、不堵车还能顺路买杯咖啡的路线”APP立刻给出了精准方案。最近，上海交大的研究团队推出的PathGPT，用LLM彻底刷新了对导航的认知。论文链接：https://arxiv.org/abs/2504.05846以前的导航算法，比如Dijkstra最短路径算法，就像个认死理的机器人：最短距离就是王道，别的我不管。但现实中开车、走路哪有那么简单？有人喜欢绕路看风景，有人急着赶路要避开拥堵，还有人想顺路接孩子。这些复杂需求，传统算法根本搞不定。后来有了机器学习模型，能从历史轨迹数据里学规律，比如根据交通状况推荐路线。但它们也有硬伤：训练好的模型只能按固定规则工作。想新增“经过某个商圈”的需求？得重新训练一个模型，费时费力还烧钱。PathGPT的核心思路：既然大模型（如GPT）能理解自然语言，那为啥不让它来翻译用户的需求？比如用户说“最快路径”“风景优美”，大模型能听懂，还能结合历史路线数据生成方案。但LLM也有短板：它可能不知道某个城市的具体道路情况，甚至出现幻觉。所以PathGPT加了个“外挂”：检索增强生成（RAG）技术。简单说，就是先建一个历史路线知识库：把每条路线的起点、终点、经过的道路名称都转换成自然语言描述，比如从人民广场到外滩，经过南京路、中山东一路。当用户提问时，PathGPT先从知识库中找出和用户起点、终点相似的历史路线。实测效果：灵活度拉满，还有进步空间研究团队在北京、成都、哈尔滨等城市的出租车轨迹数据上做了实验。结果显示，PathGPT在生成“最快路径”时，精确率和召回率虽然比一些传统机器学习模型略低（比如在哈尔滨数据集上，最快路径精确率48.4%），但它有个绝杀技——能处理那些没提前训练过的奇葩需求。比如用户提出要走一条经过三个公园的路线，传统模型可能直接懵圈，但PathGPT能结合自然语言和历史数据，给出合理方案。PathGPT的最大意义，是让导航变成自然对话。以后导航，直接像跟朋友问路一样说话就行。当然，它还有改进空间，比如偶尔会指错路（LLM的幻觉问题），在复杂路况下的可靠性还得提升。但想想看，大模型才发展没多久，就能做到这一步，未来要是结合更精准的实时交通数据、更完善的知识库，说不定真能成为每个人的智能出行管家。ChatGPT不光会聊天，还能靠实时数据给实用建议，传统地图失灵时特别管用。AI正在改变野外探险的方式，靠坐标和逻辑在没图的森林里导航，绝对是里程碑。","kind":"news","is_publish_news":true,"is_publish_highlight":false,"is_publish_live":false,"is_publish_wemedia":null,"editions":null,"column":"","sentiment":"0","news_tag":"","news_rank":0,"symbols":[],"gpt_button":0,"need_auth":false,"code":"91000000","status":"200"},"commentList":[],"isCommentEnd":true,"newsSizeData":{"likeSize":0,"commentSize":0,"repostSize":0,"favoriteSize":0,"likeStatus":false,"favoriteStatus":false},"APP":{"userAgent":"Mozilla/5.0 AppleWebKit/537.36 (KHTML, like Gecko; compatible; ClaudeBot/1.0; +claudebot@anthropic.com)","isDev":false,"isTTM":false,"deviceId":"web-server-community-laohu8-v3","version":"4.35.12","shortVersion":"4.35.12","platform":"web","vendor":"web","appName":"laohu8","isIOS":false,"isAndroid":false,"isTiger":false,"isTHS":false,"isWeiXin":false,"isWeiXinMini":false,"isWeiBo":false,"isQQ":false,"isBaiduSwan":false,"isBaiduBox":false,"isDingTalk":false,"isToutiao":false,"isOnePlus":false,"isHuaWei":false,"isXiaomi":false,"isXiaomiWebView":false,"isOppo":false,"isVivo":false,"isSamsung":false,"isMobile":false},"href":"/m/news/2551190803","isCrawlerRequest":true}