Saturday, September 26, 2015

Are You Ready to Dance with Robots?

The world of art plays an important role in human lives. The art mesmerizes and inspires us. It unleashes the creative energy and challenges conventional thinking. It provokes new thoughts and compels us to ask new questions. Can robots play a role in the art world?

Fictional robots have been playing prominent roles in movies for many years. Star Wars movies will not be the same without C-3PO and R2D2. The use of robots in movies enables writers to create new plots and enables actors to interact with superhuman characters.

The field of robotics has made tremendous progress. We now have truly remarkable robots. Can these real robots influence the art world?

I had an opportunity to interview Huang Yi on Thursday September 24, 2015 in the Clarice Smith Performing Arts Center. He is one of the pioneers of a new form of dance. His partner is a Kuka robot!

Kogod Theater Stage (Photograph by Rebecca Copeland) 
He currently uses a large intimidating orange Kuka robot in his performances. He said that he liked the Kuka robot because of its form. He programs his “dance partner” to glide through a space in harmony with music. Huang Yi and the robot move in unison during the performance and are able to express emotions to complement and augment the ambiance created by the music. His thought provoking performance asks us to examine the relationship between humans and robots.

Huang Yi's Dance Partner
(Photograph by Rebecca Copeland)
Huang Yi likes the complete predictability of the robot moves. It makes the dance safe and enables him to keep the tempo high without worrying about the need to constantly watch the robot. Currently it takes him ten hours of programming to create one minute of performance.

I wonder how this form of dance will change as robots become more intelligent and safe? Safety will encourage many more people to explore dancing with robots. Intelligence will enable robots to react to human moves and hopefully it will become easier to create new dance moves.

Huang Yi in the lab with our Kuka robots
(Photograph by Rebecca Copeland) 
Some art students in the audience seem a bit concerned about the need to learn programming to master this new art form. Hopefully advances in the area of learning from demonstrations can eliminate this barrier.

I wonder how this art form will change if we had robots that can understand the human emotions and gauge the mood expressed by the music!

What will it take for you to dance with robots?

Monday, September 7, 2015

RoboSAM: A robot that is smart enough to call humans for help!

In my opinion, one of the most important attributes of being smart is the ability to seek help when needed. This requires realizing that help is needed and getting the right kind of help from the right source. Currently, robots do not have an ability to assess whether they can successfully complete a task or not. When instructed to do a task, they simply attempt to do it. Sometimes the task execution results in spectacular success that delights the spectators and other times it leads to an embarrassing failure that baffles everyone, except the person who programmed the robot. Clearly, if robots were to become smart, they will need to ask for help when they are unable to do a task. 

Occasional robot failures can be tolerated. However, using humans to frequently clean up the mess created by robots is simply not a viable business model for using robots. Currently, deploying robots in industrial applications requires the reliability of robotic task execution to be very high. This is accomplished by designing specialized hardware and software. Extensive system testing is needed to ensure that potential failure modes are well understood and contingency plans are developed to handle them. Typically, task execution failures shut down the line and require human intervention to clear the fault and restart the line. This type of intervention is very expensive and hence robots are not used on a task until extremely high-level reliability can be achieved. Customized hardware and software costs can only be justified if the production volume is sufficiently high and tasks are repetitive (e.g., automotive assembly lines). 

To understand the underlying challenges in robot deployment, consider the following scenario. A robot is capable of picking a part if it is presented to the robot at a certain location. However, if the part has shifted from its nominal location, the robot might not be able to grasp it. The robot does not simply know where the transition boundary between task execution success and failure lies. If the part is sufficiently distant from its expected location, as the robot attempts to grasp it, the robot might bump into it, push it further, and jam the material handling system. This can in turn trigger a system fault and shut down the system. 

In order to use robots in small production batch operations or non-repetitive tasks, we will need robots that are able to estimate the probability of task completion before beginning the task. This will enable robots to assess their own confidence in doing a task. If the robot does not have high confidence in completing a task, then it should call for help. This will enable human operators to provide the robot with needed assistance (e.g., better part pose estimation, invoking a different grasping strategy) and prevent major system faults that result from task execution failure. Please keep in mind that the human only needs to help the robot with the portion of the task that is proving to be challenging. The robot can do the rest itself. In most situations, providing task assistance help to robots is much cheaper than recovering from a system shutdown. 

My students have been building a robot to demonstrate this concept in the bin picking context. This project is called RoboSAM (ROBOtic Smart Assistant for Manufacturing). Bin picking capability is representative of a robot’s ability to perceive the desired object in the environment and to successfully pick it up and deliver it in a known pose. If the robot is not sure whether it can pick the desired part from a bin containing many different parts, then it calls a remotely located human operator for help. We call this operational concept human-on-call concept. This is fundamentally different from the human-in-the-loop concept that requires the human operator to actively monitor the manufacturing cell and take control away from the robot when the robot is about to make a mistake. The new concept requires the robot to call the human operator when it decides that it needs help. 

I believe that human-on-the-call concept is the right economic model for deploying robots. It enables humans to move away from doing boring routine tasks to do challenging tasks with which robots struggle. This model allows a single remotely situated human operator to help multiple robots on an “as needed” basis. It also enables robots to be deployed on tasks on which achieving very high success rate will be difficult. For the near foreseeable future, a large number of tasks in small and medium manufacturing companies fall in this category. 

People often ask what humans will do when robots become more widespread. In my opinion, humans will be needed to teach robots how to do different tasks and bail robots out when they are confused. The key will be to develop technologies that allow robots to ask for help when needed. Recent work in our lab is a step in that direction.

Thursday, August 27, 2015

Six Recent Trends in Robotics and their Implications

There are signs all around us indicating that the field of robotics is going through a major transformation. Robots are getting significant coverage in the media. Many big companies that have virtually nothing to do with robotics are suddenly on the buying spree to acquire robotics companies. Countries that were not on anyone’s radar screen just few years ago are emerging as major players in the robotics arena. Many designs and operational constraints associated with robots are being obliterated by the use of clouds and social media. Costs are falling rapidly, enabling new applications. The notion of what was considered a robot is changing fast. Most people now agree that drones are robots. We seem to be on the verge of something big that can hopefully impact our lives in a positive way. 

This post lists six main trends and discusses their implications. 

1. Commercial Investments: Recently the commercial sector has made significant investments in robotics. Google has bought several robotics companies. Amazon has bought Kiva Systems and morphed it into Amazon Robotics. Qualcomm has also made investments in robotics. Even venture capitalists are interested in funding robotics companies. Hopefully, these will lead to the adoption of robotics in new applications and accelerate the technology developments.

2. Emergence of New International Players: Traditionally robotics advances mostly came from Japan, the US, and a few European countries. The field is expanding and new international players are emerging. China is making significant investments in robotics. Chinese manufacturers are currently leading the world in terms of procurement of new industrial robots. They are also developing their own low-cost industrial robots. The largest commercial drone maker DJI is from China. South Korea leads the world in terms of robots deployed per 10,000 workers. Recently, South Koreans won the DARPA robotics challenge by beating teams from the US and Japan. The globalization of robotics is expected to create new opportunities and challenge the leadership of the traditional players.

3. Reduction in Hardware Costs: The cost of industrial robots and drones has been declining in the commercial sector. This is expected to enable deployment of robots and drones in new applications. The agricultural sector is being projected as a major new market for robots and drones.

4. Popularity of Drones in Civilian Sector: The use of drones in the civilian sector both domestically and internationally is expected to grow at a rapid rate. Unfortunately, these robots have major vulnerability from the cyber security perspective. Recent examples of hacking of cars illustrate the vulnerability of these vehicles to cyber-attacks. New cyber-security technologies are needed to deal with attacks that can commandeer vehicles and cause physical damage. A serious incident in this area can influence public opinion and cause a major setback for this emerging field.

5. Cloud Robotics: Robots can leverage clouds to do massive data processing and exchange information with other robots in real time. Clouds are freeing robots from computing constraints and giving robots “big enough brains” to deal with challenging situations. Advances in big data are also being embraced by the robotics community to deal with the massive data generated by sensor-rich robots.

6. Leveraging Social Media Data: Robots now have access to data on social media. They can mine data (e.g., images) on social media to gain new “perception” capabilities that can in turn expand their ability to “understand” the environment. Social media can also be used to crowd source demonstrations for helping robots acquire new skills.

Wednesday, August 26, 2015

The Impact of Social Media on Entrepreneurship

Many people have observed that social media has the potential to fundamentally change entrepreneurship opportunities. It is beginning to play a key role in many aspects of creating a new business. Facebook, Twitter, and Youtube have emerged as tools to advertise new services and products. Clearly, this is beginning to have an impact on the marketing function. LinkedIn enables startup companies to recruit talent for their new businesses. Kick Starter allows entrepreneurs to raise funds to realize their new ideas. Moreover, new companies are able to buy manufacturing services and programming help on the Internet. Amazon can help a startup in selling a new product. In today’s connected world, theoretically one can be located in a significantly remote area and yet get a product manufactured, marketed, and sold without leaving home. 

Historically, few select places such as Silicon Valley have enjoyed the reputations of being the hotbed of entrepreneurship and the birthplace of many modern technological giants. These places have provided startup companies access to funding, talent, and infrastructure to get going. Has social media reduced the impact of geographical location on the startup creation? In this post, I examine this question from the entrepreneurship culture point of view.

I currently live in the Washington DC metropolitan area. People living in DC area know that chances of a startup succeeding are very small. Many people will tell you that nine out of ten startups fail, so most students graduating from college are not willing to join a startup and instead opt for a safer option. Most of them simply do not have any one in their immediate network who has become super rich working for a startup. This lack of role models discourages them from taking perceived risks.

I often visit the Silicon Valley area to attend meetings. The culture there is very different. People seem to breathe a different air. Even though the chances of any single startup succeeding wildly are not very high, people living there strongly believe that if you simply try enough times you are bound to succeed. Many graduating students from college know people who have become widely successful by taking the startup route. The existence of these role models encourages students to join a startup. They simply do not view startups as risky ventures. Instead, not joining a startup is considered a missed opportunity.

Here is a puzzle for you. Let us assume that you are playing a game in which odds are stacked against you. The probability of you losing the game is ninety percent. What is the probability is that you win the game at least once if you try ten times? Folks living in the Silicon Valley intuitively understand the answer to this puzzle. 

Social media has impacted the functionality needed to get a startup going. However, culture is a very important aspect of getting people to be engaged in startups. Ultimately, the existence of an entrepreneurship culture is what drives people to correctly assess risks associated with the startup companies and view them as opportunities not to be missed. Social media has not yet impacted the entrepreneurship culture in a significant way. Silicon Valley continues to rule the startup world because of its well-established entrepreneurship culture.

Social media certainly has the potential to impact the entrepreneurship culture in a significant way and help free it from geographical constraints. I am not sure how long it will take for this to happen.

Friday, July 31, 2015

The Role of Advanced Manufacturing in Innovation

Here is my testimony for "Make it in America: What’s Next?" panel organized by Congressman Steny Hoyer

1. The ability to innovate will increasingly depend on the presence of a vibrant manufacturing ecosystem.

In today’s global economy, the ability to innovate is crucial to creating new business opportunities and maintaining a healthy economy. The presence of a local manufacturing ecosystem is needed to maintain the US leadership in innovation and creating new industries. Designers need to understand how the manufacturing processes work to realize innovative products that are affordable and compete well globally. This understanding is difficult to achieve if the designers unable to closely interact with manufacturing engineers and experience manufacturing first hand. In today’s fast-paced world, designers need rapid access to manufacturing processes to try many different concepts to select the winner. Often the understanding of innovations in manufacturing processes can also lead to innovations in products. A nation cannot simply hope to continue to be at the forefront of innovations without having a healthy manufacturing infrastructure. There are many nations that are aggressively competing with the US in the innovation arena. The US has done remarkably well in leading the world in providing ground-breaking innovations. Many of these innovations came from companies located in Maryland. But the past performance alone cannot ensure continued future success. The US should make every effort to ensure that it maintains a healthy manufacturing sector.

2. A healthy manufacturing sector provides well-paying jobs and is crucial to the national security.

In addition to enabling innovation, a healthy manufacturing sector is necessary to provide well-paying jobs and maintaining favorable employment numbers. High-value manufacturing also creates export opportunities and helps with the trade balance. A healthy manufacturing sector is also needed to ensure national security. We should never be in a position to import parts that are critical to national security. In today’s era of constant cyber threats, we do not want to become vulnerable by importing parts that might have intentionally placed malware or serious security loopholes. Not doing so will simply give an opportunity to our adversaries to neutralize our technological superiority.

3. Recent advances in manufacturing are creating new opportunities for the US in high-value manufacturing.

The field of manufacturing is currently undergoing major changes. 3D printing is expected to revolutionize manufacturing. It enables designers to realize complex designs rapidly. The cost of 3D printers is dropping dramatically. This means that people who did not have access to manufacturing until now can buy 3D printers and make things themselves. Recent advances in robotics are reducing the need for manual labor and hence making manufacturing economically viable in high-wage rate regions. The Internet of Things technology is expected to lead to smart manufacturing. Companies need to offer high quality products of increasing complexity at a faster pace with lower prices. This makes manufacturing very challenging. Smart manufacturing technologies are expected to significantly improve manufacturing efficiency and productivity. These technologies can also be used to reduce negative environmental impact of manufacturing. Recent advances in materials such as digital materials, multifunctional materials, metamaterials, and programmable materials are expected to enable a new generation of products. Almost all of these advances originated in the US. We should leverage these advanced manufacturing technologies to grow manufacturing industry in the US and Maryland.

4. Advanced manufacturing will require a workforce with strong STEM background.

Advanced manufacturing requires a different kind of workforce. Rather than relying on manual skills, people are expected to work with sophisticated machines. The nature of the products is also expected to change rapidly. This requires a very different kind of workforce. Training the workforce for the next generation manufacturing technologies will require a strong emphasis on STEM subjects and a new pedagogical approach. Schools, colleges, and universities will need new labs with access to advanced manufacturing technologies. A closer partnership with industry will also be needed to ensure that the workforce training programs match the skills required by the industry. The University of Maryland is developing new labs and courses in the Advanced Manufacturing area to support Maryland-based businesses.

5. Recent technological advances are expected to create new business opportunities.

New markets and industries will be created around several emerging areas such as unmanned systems, driverless cars, electric vehicles, next generation batteries, intelligent prosthetic devices, smart appliances, and personalized medicine. Many of these technologies were developed in the US. The US should strive to become the leading world manufacturer and exporter of products in these emerging areas. The state of Maryland should lead the nation by demonstrating how to leverage recent innovations to launch new manufacturing-based businesses.