标签: car

5月13日，滴滴出行宣布了一则消息，其最新一轮融资取得了重要进展，苹果公司将以10亿美元投资本轮融资。这是滴滴创办至今获得的单笔最大投资数目。通过这次投资，苹果也成为了滴滴战略投资者。 滴滴出行创始人兼CEO程维表示：“Apple的认可令成立四年的滴滴深受鼓舞，更是一种激励。 滴滴将继续努力，与司机和乘客、与全球伙伴一起，让人人拥有更加灵活可靠的多元出行选择；帮助我们的城市解决交通、环保和就业挑战。” Apple CEO Tim Cook则表示：“滴滴彰显了中国iOS开发者社群的锐意创新精神。滴滴打造的出行平台及其卓越的管理团队令人赞叹，我们期待支持滴滴的长期发展。” 作为世界科技行业的领先公司，苹果如此大手笔在中国科技公司身上投资尚属首次，苹果为何要在中国投资？为何要投滴滴？两者合作的基础与前景如何呢？我们来看一下。   一、苹果为何要在中国投资？ 提起苹果，我们相当的往往是乔布斯，是一家提供优质科技产品的科技公司。而鲜为人知的是，苹果的财务操作也是高手。 美国的一项调查显示，苹果公司的避税手段非常高明。如果把苹果公司的海外资产带回美国，美国税收部门将可对其征税约592亿美元，而苹果避税的一个重要方式就是在海外投资。 美国企业最高所得税为35%，在全球排前列。按照美国税法，居民公司（在美国境内根据美国法律组成或创建的公司）就其来源于全世界的所得纳税。但是美国税法中的一些例外条款规定，只要海外的利润用于再投资、没有返回美国国内，该公司就可以暂时不为这部分利润缴税。 中国作为苹果全球最大的市场，多年来积累了大量的利润，这些利润如果不在中国投资，而是返回美国，就会被征收35%的所得税。如果有好的途径，做一个投资当然是理智的选择。所以苹果要那中国获得的利润在中国投资。 除了避税，苹果与中国政府之间的关系也很微妙。前不久。按照中国政府对网络发布内容、特别是对外资企业发布的内容实施了严格的监管。而4月份苹果的iTunes电影和iBooks商店就被封。这引发了苹果投资者的担忧。 美国注明的激进投资者伊坎就为此抛空了持有的4580万股苹果股票。 近期，库克将再次访华，不知是否与此事有关。不过按照美国人对中国政府公关的理解，在当地投资被视为友善的举动。 微软就曾经在中国大肆投资，获得了良好的政府关系，促进了微软产品在中国的推广。 而高通在遭遇反垄断调查，付出罚款之后，立即投资了中芯国际合作新技术，还把订单给了中芯国际。而英特尔更多多年来持续与中国科技公司合作。 所以，苹果这次在中国投资一方面可以财务避税，转移在中国的巨额利润，一方面还能搞好与政府的关系，为未来的商业计划铺路。可谓一举两得。 二、苹果为何要投资滴滴 滴滴成立已经数年，其发展有一个明显的策略。 首先，滴滴补贴出租车司机，培养人们用APP叫车的习惯，在司机中普及APP，培养大量用户。 然后，滴滴开始开放专车，绕开出租车，给专车司机补贴，形成专车与出租车的竞争。 而在政府废除出租车管理办法，网约车管理办法即将出台的时候，滴滴又开始引入自营约租车，降低补贴，逐步淘汰非专职的私家车。 通过三步走的计划，现在滴滴已经掌控了中国的网约车市场，将获得运营权，分享了原本属于各个城市出租车公司的蛋糕，可以名正言顺的收取服务费盈利。 滴滴已经过了烧钱补贴的阶段，进入到了收获的阶段。而苹果的进入恰是时候。 这次10亿美元给滴滴，滴滴不用去烧掉。而发展起来的业务是真正可以代来利润的业务。 当然，滴滴本身也不简单。数据显示，滴滴已经在中国拥有绝对领先的位置。覆盖400个城市，其中200个城市达到盈亏平衡点，日订单1100万，覆盖出租车、专车、顺风车、代驾、试驾、巴士、企业级等丰富的产品线，占据着中国网约专车87%的市场份额。2015年滴滴出行完成订单14.3亿，相当于整个美国移动出行市场的2倍。仅北京一个城市，滴滴的日订单量就相当于纽约每日全部移动出行量的6-7倍。 按照市场研究公司CB Insights数据，滴滴快的已经成为继Uber、小米、Airbnb之后的全球第四大最有价值“独角兽”。 按照1100万的订单计算，滴滴收一定比例服务费，很容易日进亿金，堪比中国移动。成为滴滴的投资者，即使没有IPO，只是吃红利也会获利丰厚。 所以苹果投资滴滴是可以吃蛋糕的。 此外，苹果正在搞苹果汽车，苹果汽车从目前的消息看是电动汽车，很可能会有无人驾驶功能。 电动汽车相比燃油车的优势是单位里程费用较低，对于私家车这个优势不明显，而对于一天几百公里的出租车，这个效益是很显著的。滴滴很有可能成为未来苹果汽车的下游客户。 同样，无人驾驶技术最大的效益也是用于替代成本高昂的出租车司机，无人驾驶技术实用化之后，苹果与滴滴合作，也可以分享一块大蛋糕。 而且，滴滴还有苹果没有的技术。 滴滴出行是一个移动的智能出行网络。包括随区域、时间变动的定价、订单的高效匹配、根据供需预测之后的司机运力调度等等。对于一个具体区域来说，滴滴“数据大脑”已经实现了提前15分钟时间达到超过88%准确率的预测，合理调度司机运力，缓解拥堵。 滴滴出行平台上每天的计算次数以10亿计，高峰时段，每一分钟的匹配就高达200万次；每天平台都要计算所有司机和订单的加乘；每天平台上新增数据50TB以上，每天路径规划次数达90亿次。 苹果要做无人驾驶，滴滴出行的数据都是宝贵的基础数据，可以用于无人驾驶系统对路线的深度学习，中国道路的实际大数据是谷歌都拿不到。 所以在技术上苹果与滴滴合作也非常划算。 最后，苹果的车联网平台CarPlay在于滴滴合作也变得更容易普及。 多种因素之下，苹果选择了滴滴。 三、你情我愿，前景美好 目前滴滴正在进行新一轮融资，这一轮融资规模据说达到20亿美元，而滴滴的估值据说已经达到了250亿美元的级别，俨然一家世界巨头，而这还是如今资本寒冬的数据，含金量远高于泡沫时期的什么450亿美元。 这次有苹果这种顶尖的科技公司投资，滴滴也是非常高兴的。除了财务上融资。苹果能给滴滴的帮助还有很多。 一方面，iPhone本身是滴滴的一个入口，而且是一个全球化的入口。这对滴滴的全球化非常有利。在中国滴滴已经拿下87.5%的市场，而全球滴滴还有很广阔的空间，滴滴未来是要与Uber争夺世界第一的。苹果的加盟让滴滴如虎添翼。 此外，未来的苹果汽车对滴滴也是很大的诱惑。约租车的主要成本一是汽油，二是司机的工资。而苹果未来的无人驾驶电动车恰恰可以解决这两个问题。苹果汽车本身的品牌也可以帮助滴滴拓展业务。要知道，特斯拉的约租车基本是供不应求的。 所以，滴滴很高兴的拿了苹果的10亿美元投资。 综合来看，这次投资算是双赢，苹果一举多得，滴滴也是一举多得，而两者合作后的前景更是美好。 滴滴可以在全球扩展业务，用未来的苹果汽车降低运营成本；而苹果先是做了一笔划算的财务投资，可以很快就有漂亮的财报。而后还给自己未来的产品找到一个全球化的大买主，至少首批Apple Car的销量不用担心了。而滴滴做的越大，越全球化，苹果汽车未来的需求就越大，苹果可以吃到投资的收益，还能吃到未来苹果汽车销售的利润。这无疑是一笔划算买卖。 库克确实是一个精明的商人。  

Today's internal combustion (IC) and EV/HEV cars are already pulling kilowatts from the battery and power subsystem, and the drain is clearly going to increase. The power train control, infotainment, ADAS (advanced driver assistance systems), and other functions each need "juice." Ten-kilowatt load factors are a very real possibility.   The problem is that the standard 12.6 V (nominal) supply rail brings an inherent limitation. Since P = VI, simple math shows that each kW of demand will need about 80 A, and that's a serious amount of current. As the current drawn from the source increase, resistive voltage losses increase linearly as defined by V = IR, but dissipation increases with the square of the current: P = I2R. The solution to the loss challenge has been known since the early days of power (think Edison, Tesla, and Steinmetz) use higher voltages to reduce the current needed when delivering a given amount of power.   It's the same reason that many homes and offices have 240 VAC mains for some equipment (electric dryers, EV chargers) in the U.S. instead of 120 VAC alone, and why industrial factory mains are often 480 VAC. To minimize the supply inefficiency dilemma, there's talk of migrating cars to a 24-V or even a 48-V rail. A recent article at IHS Engineering360, " 48 Volts: Time for a Jolt to Vehicle Performance? ,”explored the promise, potential, and even peril of higher voltages in cars.     This is not the first time that there has been a push to using higher voltages in cars. About a decade ago, there was some effort to go to 24 V or 48 V, for the same reason that 48 V is now being promoted -- except the problem wasn't as severe or imminent as it is now. That 24/48-V effort failed for many reasons, but the underlying rationale on dealing with the physics constraints versus the load demands are unchanged. If it happens at all, the transition to 48 V will not be easy. The reality is that there is a significant ripple effect from such an apparently modest, easily defined change to a long-established standard. Increasing the nominal voltage affects cabling, connectors, DC/DC converters, small-motor drivers, regulatory concerns, fusing, failure modes, test and maintenance, dealer support, and more. It's a teachable moment to keep in mind when you hear people who don't understand system and product design casually suggest, and with confidence, "what's the big deal? Just go to bigger batteries."   Regardless of the voltage rail selected, there is another limit on the electrical subsystem, only this time at the source rather than the current-delivery function. The car battery is not only a source of voltage, it is an energy-storage unit, and a standard lead-acid battery has density of about 40 W-hr/kg and 100 W-hr/liter.   Even if you increase the voltage supplied by the battery by using more cells, there will still be issues related to how much energy and power (the rate at which energy is being used) a battery can actually deliver from a given volume, and if the alternator can provide energy and power to keep charging that battery. While non-lead acid battery chemistries (such as lithium-based ones) have much higher densities, the venerable lead-acid has some very favorable attributes (cycles, self-discharge, temperature performance) which make it a better fit in many ways for the basic task of starting the car and keeping its rails powered.   Even EV/HEVs, with their high-voltage/power battery assemblies, are not immune to problem of powering all these other loads. You don’t want to be running that much-higher voltage bus around the car just to power accessories, nor do you want to drain the traction motor's batteries power them, among many other practical issues.   What's you view on the desirability and practicality of going to 48-V battery packs and subsystems in both IC and non-IC cars? Will it happen this time, due to the inevitable pressures of more loads? Or are the associated problems going to make it a great idea in principle, but an impractical one in practice?

I check my car tyres about once a month just like many people, since correct pressure is necessary for good car handling, a smooth ride, and good gas mileage. When I checked my front tyres recently with my Topeak digital gauge (image below), one was at the correct pressure (30 psi), but the other was higher, at about 33 psi. What puzzled me was that they had both been at 30 psi the previous time. I know tyres can lose pressure, but I had never heard of a case where the pressure increased on its own.     This inexpensive digital-pressure gauge is a pleasure to use: It reads up to 160 psi/11 bar (useful for bike tyres and suspensions) to three significant figures; can be switched among psi, Bar, or kg/cm² readings; and handles Presta and Schrader valves -- a big improvement over the old "pencil-type" mechanical tyre-pressure gauges.   I gave this some thought and saw only two possibilities at first: Someone was playing a practical joke on me (very unlikely); or my previous reading for just that one tyre alone was in error (also unlikely, as all tyres were measured twice, and at the same time).   Then I looked at the car and saw that the tyre that read high was in full sunlight, while the other was in the shadow, and a black tyre certainly does heat up from solar radiation. Mystery solved -- or maybe not. I pulled out my custom-made "back of the envelope" pad and did a quick calculation using the ideal gas law :   P × V /T = K or P = K × T / V   ...where P is pressure, V is volume, T is the absolute temperature, and K is a constant, which depends on the amount and type of gas (here, the value is irrelevant).     My hand-made "back of the envelope" pad reminds me that doing quick, rough estimates is often a good first step to understanding the parameters of a problem.   Thus if the pressure I measured was about 10% higher than the original, and the volume was constant, then the temperature of the air in the tyre also should have gone up about 10%. The "cold" ambient temperature was about 77°F (25°C) or about 300K (remember, this is a rough estimate we're doing), so the delta rise would be 30K (30°C), or about 55°F.   Then I worried that perhaps a change in the tyre's volume would affect my estimate, but I realized it was a non-issue for two reasons. First, a car tyre is not an easily expandable balloon; it is a rubber enclosure restrained by steel-wire belts. Therefore, its volume stays fairly constant, especially for modest variations around a nominal value (this is a type of assumption we often use in many simplified models).   Second, even if the tyre did expand slightly due to the increase in internal pressure, that would actually cause a decrease in the resultant pressure -- again, the gas law. (I recall seeing a complex differential equation embodying the relationship among a tyre's construction, pressure, and volume, for more advanced modeling.)   Was solar heating the answer to my mystery? I don’t know, as I have no way of measuring the internal air temperature. I suppose I could do some thermal modeling, or even use an application such as COMSOL Multiphysics for a thermal/mechanical simulation, but it's not worth the effort.   So the question of sunlight heating the tyre and raising the pressure remains a slightly open mystery. My "gut" tells me that a 30°C/55°F rise for a black-rubber tyre in full sunlight is possible, but that's where I have to stop.   Do you think it was solar-heating effect? Can you think of any other causes? Have you ever had a similar "simple" measurement mystery, where you are not sure of the actual cause of the observed effect?

   当前许多厂牌（包含TI, ADI, NXP, Averlogic等等）的影像译码IC(video decoder)可以将影像的CVBS讯号转换到数字的656 bus上,提供给后端的处理器使用. 一个优良的Video Decoder需要内部ADC的妥善处理。从早期的8~9bit 处理到现在10~12 bit处理,可以将影像的画质做的更好的分解运算,同时在低耗电的制程上,新一代的Video Decoder可以具备更好的Power Saving mode, 低电压的制程使系统的模拟电路更为简洁,耗电量更低于一般的Video Decoder. 例如Averlogic AL240 与NXP SAA7113作比较,可以清楚的看出差异性,请参阅Averlogic网站作查询.

现在有许多的场合需要用到多画面显示,例如大卖场,百货公司大门监控,大楼监控,学校,军事基地,银行,机器人视觉,高速公路,医院等,这些都需要一个高性价比的处理器. 目前TI, ADI等厂商都有类似产品,台湾厂商AVERLOGIC 也有类似产品, 在ZOOM IN,ZOOM OUT, video scaling, OSD, multiple window, PIP, POP,Video multiplexing, 4/9/16 channels window mode, OSD built in,这些项目都可以做更有弹性的开发设计. AVERLOGIC AL37204似乎可以满足需求: Video Picture in Picture Video Picture over Picture Video capture, Video scaling up and video scaling down Multiple view windows 4/9/16 channels video out for VGA/CVBS out. Video loss/blind detection. Motion detection. 在车用,居家场合上都需要有一个经济型的,简单设计的应用,AL37204基于8051,Keil C的应用开发环境可以满足大多数场合的使用.请各位博客提供建议! 联系人： 龚先生（茂晶骏龙有限公司） 电话：86755-8828-5788 邮箱：DragonG@gfei.com.hk