Global Economic
Trend Analysis

Recent Posts

Thursday, March 26, 2015 3:12 PM

For Robots Only: Amazon Sponsored Contest; Soft Fingers Needed

Amazon is sponsoring a robot warehouse automation contest to see to who can pack the most boxes in the least amount of time without dropping any packages or crushing anything delicate such as cookies.

In the contest, in which human workers are not eligible to apply, the robots will have to work without any remote guidance from their creators.

Please consider the MIT Technology Review, Amazon Robot Contest May Accelerate Warehouse Automation.

Packets of Oreos, boxes of crayons, and squeaky dog toys will test the limits of robot vision and manipulation in a competition this May. Amazon is organizing the event to spur the development of more nimble-fingered product-packing machines.

Participating robots will earn points by locating products sitting somewhere on a stack of shelves, retrieving them safely, and then packing them into cardboard shipping boxes. Robots that accidentally crush a cookie or drop a toy will have points deducted. The people whose robots earn the most points will win $25,000.

Amazon has already automated some of the work done in its vast fulfillment centers. Robots in a few locations send shelves laden with products over to human workers who then grab and package them. These mobile robots, made by Kiva Systems, a company that Amazon bought in 2012 for $678 million, reduce the distance human workers have to walk in order to find products. However, no robot can yet pick and pack products with the speed and reliability of a human. Industrial robots that are already widespread in several industries are limited to extremely precise, repetitive work in highly controlled environments.

Pete Wurman, chief technology officer of Kiva Systems, says that about 30 teams from academic departments around the world will take part in the challenge, which will be held at the International Conference on Robotics and Automation in Seattle. In each round, robots will be told to pick and pack one of 25 different items from a stack of shelves resembling those found in Amazon’s warehouses. Some teams are developing their own robots, while others are adapting commercially available systems with their own grippers and software.

The challenge facing the robots in Amazon’s contest will be considerable. Humans have a remarkable ability to identify objects, figure out how to manipulate them, and then grasp them with just the right amount of force. This is especially hard for machines to do if an object is unfamiliar, awkwardly shaped, or sitting on a dark shelf with a bunch of other items. In the Amazon contest, the robots will have to work without any remote guidance from their creators.

“We tried to pick out a variety of different products that were representative of our catalogue and that pose different kinds of grasping challenges,” Wurman said. “Like plastic wrap; difficult-to-grab little dog toys; things you don’t want to crush, like the Oreos.”

While the Amazon challenge might seem simple, Saxena believes it could quickly make an impact in the real world. “If robots are able to handle even the light types of grasping tasks the contest proposes,” he says, “we could actually start to see a lot of robots helping people with different tasks.”
The preceding MIT review describes the 2015 ICRA Contest May 26-30 in Seattle.

2014 Participant Video

The above video shows the 2014 University of Colorado Amazon Picking Entry.

The Baxter robot actually seems a bit clumsy because grasping and packing random objects is not a precisely repetitive action.

The University of Colorado video gets interesting towards the end, highlighting advancements in "soft fingers" and human-like skin.

Meet Baxter

I discussed Baxter on January 22, 2013 in Meet "Baxter" the Robot Out to Get Your Minimum-Wage, No Benefits, Part-Time Job, Because He's Still Much Cheaper; Fed Cannot Win a Fight Against Robots.

Soft Fingers Needed

Soft fingers and better motions are needed to win the 2015 contest, and they are coming (image from the University of Colorado video).

Anything that can be automated, will. The higher the minimum wage and the lower the interest rate, the more incentive companies have to replace humans with hardware and software robots.

Mike "Mish" Shedlock

1:58 PM

Damn the Reports, Full Speed Ahead; Recession Overdue; Good Time to Normalize Rates?

Here's one for the I'll believe it when I see it category: Fed Officials say Rate Hike Plan Intact Despite Weak U.S. Data.

In separate events in Frankfurt and Detroit, St. Louis Fed President James Bullard and Atlanta Fed President Dennis Lockhart said U.S. monetary policy might need to be adjusted in light of the economy's steady improvement since the 2007-2009 financial crisis.

"Now may be a good time to begin normalizing U.S. monetary policy so that it is set appropriately for an improving economy over the next two years," Bullard said at a conference in the German financial hub.

The challenge now, Lockhart said, is to sort out whether recent weakness in exports, manufacturing and capital investment indicate the start of an economic slowdown or other temporary factors such as the soaring value of the U.S. dollar.

Lockhart said he is confident for now that the weakness is "transitory," and still regards it as highly likely that the Fed will raise rates at either its June, July or September meetings.

"We're still on a solid track ... The economy is throwing off some mixed signals at the moment and I think that is going to be passing or transitory," Lockhart said in an interview with CNBC from a Detroit investment conference.

"In the beginning when the dollar declined I was prepared to, to some extent, dismiss the influence of the dollar as being not great because our economy is not so export-dependent, but I'm upgrading it as a factor to watch," he said.
Totally Clueless

In simple terms, Lockhart may as well have said that he is "totally clueless."

We are going on 7 years of economic expansion.

The San Francisco Fed has an interesting report on the Duration and Timing of Recessions.
NBER records show that, over the period from the mid-1940s until 2007, the average recession lasted 10 months, while the average expansion lasted 57 months, giving us an average business cycle of 67 months or about 5 years and seven months. However, there has been considerable variation in the length of business cycle expansions and contractions in the past.

The shortest recession between the mid-1940s and 2007 lasted only six months, from January to July 1980. The two longest recessions during the period lasted 16 months each, one extending from November 1973 to March 1975, and the other from July 1981 to November 1982. In both of these periods there was a noticeable decline in real GDP.

In contrast to the relatively short duration of most recessions, periods of expansion tend to last much longer, helping the economy expand over time. The shortest expansion period from the mid-1940s until 2007 lasted only 24 months, from April 1958 to April 1960. The longest expansion continued from March 1991 to March 2001, setting a record of 120 consecutive months of growth.
Recession Overdue

Statistically speaking, a recession is overdue although there is wide variance in both the length of recessions and recoveries.

Yet, there is very little reason to believe weak report after weak report is "transitory". The idea "we're still on a solid track," is downright ludicrous.

Good Time to Normalize Rates?

Is this a good time to normalize rates?

Let's answer it this way: It's better than a month from now but not as good as two years ago. In fact, for the second time, rates never should have gotten as low as they did for as long as they did.

The Fed has sponsored three asset bubbles in recent history, each of increasing amplitude.

Three Major Bubbles

  1. Dot-Com bubble
  2. Housing and credit bubble
  3. Global equity and junk bond bubble

Bubble number three is still expanding. Few admit that it's a bubble, simply because it  hasn't popped yet.

If the Fed does hike (which is doubtful because Yellen is calling the shots, not Bullard or Lockhart), most will point a finger and say the "Fed caused a needless recession".

Nothing could be further from the truth.

By blowing yet another asset bubble, the Fed guaranteed another hugely destructive asset deflation bust.

Why hike? The reason to hike is the bigger the bubble, the bigger the bust, something the Fed should have thought about in advance but didn't, and never does.

Mike "Mish" Shedlock

2:17 AM

Jobs and Employment: How Much Recession Warning Can One Expect?

Watching the Wrong Things

Many market watchers have their eye on jobs and the unemployment rate as the determinant of when the Fed will hike.

Let's investigate the wisdom of that approach with actual data.

I downloaded seasonally adjusted employment and jobs data for the last five recessions from the BLS. Because the recessions start in different months, use of seasonally adjusted data is mandatory for this exercise.

My focus is on jobs and employment in the period three months prior to the recession to three months after the recession.

I used the National Bureau of Economic Research (NBER) report on US Business Cycle Expansions and Contractions as the official arbiter as to when recessions begin.

Jobs are from the Establishment Survey. Employment is from the Household Survey. Results are similar.

Recessions vs. Employment (in Thousands)

Recession DateEmploy -3 MonthsEmploy -2 MonthsEmploy -1 MonthsEmploy -0 MonthsEmploy +1 MonthsEmploy +2 MonthsEmploy +3 MonthsPeak Employment
1980-0199,40499,57499,93399,87999,99599,71399,2331 Month Later
1981-07101,056101,048100,298100,693100,689100,064100,3781 Month Later
1990-07118,852119,151118,983118,810118,802118,524118,5361 Month Prior
2001-03137,614137,778137,612137,783137,299137,092136,873Recession Start
2007-12146,244145,946146,595146,273146,378146,156146,0861 Month Prior

Recessions vs. Jobs (in Thousands)

Recession DateJobs -3 MonthsJobs -2 MonthsJobs -1 MonthsJobs -0 MonthsJobs +1 MonthsJobs +2 MonthsJobs +3 MonthsPeak Jobs
1980-0190,48290,57690,67390,80290,88290,99490,8502 Months Later
1981-0791,28391,29391,49091,60291,56691,47991,380Recession Start
1990-07109,687109,839109,862109,834109,613109,525109,3661 Month Prior
2001-03132,723132,696132,767132,742132,460132,422132,2911 Month Prior
2007-12138,053138,136138,253138,350138,365138,278138,1991 Month Later

How Much Warning Can One Expect?

The answer is clearly none.

In the previous five recessions, jobs peaked two months after the start of the recession once, one month later once, one month prior twice, and once during the recession month.

In the previous five recessions, employment peaked one month after the start of the recession twice, one month prior twice, and once during the recession month.

The NBER says the last US recession started in December of 2007 and lasted until  June of 2009. Let's take a closer look at stats from that recession.

2007-2009 Recession Stats

  • Jobs were higher three months after the recession started than two months before the recession started.
  • Jobs were higher two months after the recession started than one month before.
  • Employment was higher three months after the recession started than two months before.
  • Employment was higher two months after the recession started than one month before.

In the last recession, jobs and employment did not provide a clear signal for months.

Negative Data Pours In

With the exceptions of jobs, most other data has been negative.

Economists keep watching jobs, a lagging indicator, somehow convinced that jobs tell the story of what the Fed is about to do. Such a focus is complete silliness. When it's clear that jobs have turned, the economy will likely be in recession.

Mike "Mish" Shedlock

Last 10 Posts

Copyright 2009 Mike Shedlock. All Rights Reserved.
View My Stats