{"id":9996,"date":"2021-02-28T17:45:12","date_gmt":"2021-02-28T16:45:12","guid":{"rendered":"https:\/\/advancedfleetmanagementconsulting.com\/eng\/?p=9996"},"modified":"2021-02-28T17:45:12","modified_gmt":"2021-02-28T16:45:12","slug":"driver-data","status":"publish","type":"post","link":"https:\/\/advancedfleetmanagementconsulting.com\/eng\/2021\/02\/28\/driver-data\/","title":{"rendered":"Robots Need Human Driver Data to Succeed, Says Wipro"},"content":{"rendered":"

Everyday driver data is the best way to advance autonomous driving technology rather than persisting with sterile robot driving on-road testing.<\/span><\/p>\n

That\u2019s the view of\u00a0 Thomas Mueller, CTO engineering and global practice, NextGen Services at Wipro. He points to Tesla as the leader in harvesting vital autonomous driving data from its customers\u2019 vehicles, placing it at the cutting edge of future driverless capabilities.<\/span><\/p>\n

Mueller suggests data is expanding autonomous driving beyond the confined testing environments that they have at the moment. He said: \u201cThe thing is the nature of artificial intelligence systems (AI) is that you need to feed them relevant training data and to do this at scale it requires you to be in many markets to acquire that data for the specific situations that your teams and algorithms consider valuable. For example, if you want to figure out how to handle a situation where a police car with blue lights on was parked in a market like Italy, you have to be able to acquire a reasonable amount of training material that can feed the algorithms the right sort of information. How you do this is to this with the traditional way of a test fleet of human drivers but this is not a scalable model.\u201d<\/span><\/p>\n

Mueller says the industry has wasted too much time using data harvested from robots and not the relevant data taken from human drivers. He explained: \u201cSo the industry has spent many years doing this model and, hence, the algorithms are utterly stupid. To make the robotic of synthetic driver to behave like a human, it requires them to watch the human driving the car for a very long time at scale. Unfortunately, from an infrastructure point of view, there is only one company that has managed to do this at scale is Tesla.\u201d<\/span><\/p>\n

He pointed to Tesla\u2019s decision to place high powered computers in its cars whose main role was to collect driver behavior information. Mueller explained: \u201cIts new computer does all the processing in the glove box while people just drive the car. Giving the car owner a computer like this for free was a very bold decision along with all the other sensors. It\u2019s an investment that Tesla made to help it develop the differentiating software that gives it these [autonomous] capabilities. So, Tesla has built the car as a data engine so it can send detection notes to the whole fleet continuously to figure out certain valuable learning situations.\u201d<\/span><\/p>\n

Fatal data<\/strong><\/span><\/p>\n

It does raise the question about whether the multiple serious and deadly crashes the owners have suffered while using Tesla\u2019s autonomous features will also be part of this \u2018valuable\u2019 learning? However, Mueller said the value of human driver behavior will always outweigh that of the robot. He added: \u201cThis delta data of comparing a human driver who is actually driving day-to-day and comparing it to what the synthetic driver, which is getting smarter by the day, is the most valuable data there is. With a test driver sitting in an autonomous vehicle and doing nothing, the computer cannot compare how a human can handle a situation. So it will always behave like a robot and not like a human.\u201d<\/span><\/p>\n

Now he says it\u2019s up to other carmakers to catch up and the best way would be to begin sharing their own driver data with each other. Mueller said: \u201cThe biggest problem is that carmakers are not sharing autonomous data with each other. This is almost stupid because it\u2019s prohibitive to really leverage the data to get better software quality. This also just the testing data and not the everyday driver data. The ability to have the everyday driver contributing to the data, that might start this year with the latest BMW architecture, would improve data quality.\u201d<\/span><\/p>\n

He concluded that the market value of Tesla could be justified because of the vault of driver data it has built up. He said: \u201cIf Daimler would have held on to its stake in Tesla, the holding would be worth more than the entire company is today! So instead of making cars they could have just held on to their investment in Tesla, they would have done better for their shareholders.\u201d<\/span><\/p>\n

 <\/p>\n

\u2014\u00a0Paul Myles is a seasoned automotive journalist based in London. Follow him on Twitter\u00a0@Paulmyles_<\/a><\/em><\/p>\n

\"Paul<\/div>\n
\n

Paul Myles<\/a><\/h3>\n

TU-Editor<\/p>\n<\/div>\n

 <\/p>\n

Source: https:\/\/www.tu-auto.com<\/a><\/span><\/p>\n

CUT COTS OF THE FLEET WITH OUR AUDIT PROGRAM<\/strong><\/a><\/h3>\n

\"\"<\/a><\/p>\n

The audit is a key tool to know the overall status and provide the analysis, the assessment, the advice, the suggestions and the actions to take in order to cut costs and increase the efficiency and efficacy of the fleet. We propose the following fleet management audit.<\/p>\n

FLEET MANAGEMENT AUDIT<\/strong><\/a><\/h3>\n","protected":false},"excerpt":{"rendered":"

Everyday driver data is the best way to advance autonomous driving technology rather than persisting with sterile robot driving on-road testing. That\u2019s the view of\u00a0 Thomas Mueller, CTO engineering and global practice, NextGen Services at Wipro. He points to Tesla as the leader in harvesting vital autonomous driving data from its customers\u2019 vehicles, placing it…<\/p>\n","protected":false},"author":3,"featured_media":9997,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":[],"categories":[18],"tags":[39],"_links":{"self":[{"href":"https:\/\/advancedfleetmanagementconsulting.com\/eng\/wp-json\/wp\/v2\/posts\/9996"}],"collection":[{"href":"https:\/\/advancedfleetmanagementconsulting.com\/eng\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/advancedfleetmanagementconsulting.com\/eng\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/advancedfleetmanagementconsulting.com\/eng\/wp-json\/wp\/v2\/users\/3"}],"replies":[{"embeddable":true,"href":"https:\/\/advancedfleetmanagementconsulting.com\/eng\/wp-json\/wp\/v2\/comments?post=9996"}],"version-history":[{"count":1,"href":"https:\/\/advancedfleetmanagementconsulting.com\/eng\/wp-json\/wp\/v2\/posts\/9996\/revisions"}],"predecessor-version":[{"id":9998,"href":"https:\/\/advancedfleetmanagementconsulting.com\/eng\/wp-json\/wp\/v2\/posts\/9996\/revisions\/9998"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/advancedfleetmanagementconsulting.com\/eng\/wp-json\/wp\/v2\/media\/9997"}],"wp:attachment":[{"href":"https:\/\/advancedfleetmanagementconsulting.com\/eng\/wp-json\/wp\/v2\/media?parent=9996"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/advancedfleetmanagementconsulting.com\/eng\/wp-json\/wp\/v2\/categories?post=9996"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/advancedfleetmanagementconsulting.com\/eng\/wp-json\/wp\/v2\/tags?post=9996"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}