Please welcome back Sterling Anderson, a continuing guest blogger and Ph.D. student at MIT working in the Robotic Mobility Group. In today’s article, Sterling tells us about further developments regarding his work in semi-autonomous vehicle control. In his first post, Sterling discussed how this work relates to sustainability by helping reduce the vehicle mass required to keep its occupants safe. Here he describes an important aspect of these systems: threat assessment.
Hello again! With a few patents in the pipeline, and having given you a chance to think about what a semi-autonomous hazard avoidance system might look like, we’re ready to discuss the system we’ve developed.In this post, I briefly (re)introduce the vehicle safety problem and describe the current state of the art. I then highlight significant challenges inherent to the design of a comprehensive driver assistance system and describe the need for improved threat assessment techniques. In my next post (Part III of this series), I will point out two fundamental challenges facing existing ADAS systems and introduce our approach. Finally, Part IV will demonstrate the performance of our system and outline the challenges it faces going forward. Each part in this series will close with a thought question for the interested reader, which you are free to discuss either in the comments section below.
THE PROBLEM
Vehicular safety is a problem that, I think, needs little motivation. Recent traffic safety reports from the National Highway Traffic and Safety Administration show that in 2008 alone, over 37,000 people were killed and another 2.3 million injured in motor vehicle accidents in the United States. While the longstanding presence of collision mitigation systems (seat belts, roll cages, crumple zones, etc.) has contributed to a decline in these numbers from previous years, it has failed to eliminate collisions altogether and has limited engineers’ ability to create smaller, lighter, and more energy-efficient vehicles. This is where Advanced Driver Assistance Systems (ADAS) – systems designed to avoid hazards altogether – come in.
TYPES OF ADVANCED DRIVER ASSISTANCE
Advanced Driver Assistance Systems (ADAS) in use today can be (roughly) placed into four classes. The first of these classes can be said to be perhaps the most hands-off of the driver assistance techniques. Systems that fall into this class are broadly known as driver-warning systems and include lane departure warning systems, lane change assistance systems, and collision warning systems (sometimes imprecisely referred to as “collision avoidance systems”), among others. Driver warning systems typically provide feedback to the driver on visual, audible, or in some cases haptic (touch) channels. Research in this area is both active and complex owing largely to the immense variability that exists between human drivers and the consequent difficulty of predicting exactly how each will respond to various warning cues. For example, where a flashing light or audible tone may be helpful to some, it may unnerve, confuse, or even annoy others.
Stepping up the level of autonomy (or “hands-on’ness”) a notch brings us to collision preparation systems. This class of ADAS seeks to prepare for or help the driver avoid accidents. Examples include systems that pretension the seatbelts, prime the brakes, reduce the speed, or even adjust the suspension stiffness when a collision or threat appears imminent or when various sensors on the vehicle sense excessive wheel skid or roll angle.
One step closer to full vehicle autonomy lie Electronic stability control (ESC) systems, which help the driver avoid skidding and loss of control by selectively applying the brakes. This incredibly-useful (and increasingly-popular) ADAS class includes anti-lock brakes, yaw stability controllers, and roll stability controllers, among others.
Still greater levels of vehicle autonomy are found in what can be called “semi-autonomous” systems. The primary difference between these systems and the more passive stability control systems just described is that semi-autonomous hazard avoidance systems actively determine a course of action that may differ from the driver’s intended maneuver and may, when necessary, cause the vehicle to deviate from the course or speed that the driver’s commands prescribe. For example, the adaptive cruise control systems that since their 2006 debut in the United States have become an icon for intelligent (and expensive) high-end vehicles, determine based on the relative speed between the host vehicle (itself) and a hazard vehicle (the guy in front of you) whether, when, and how much to adjust velocity to avoid a collision. Another semi-autonomous ADAS that exists in very limited form today is the lane-keeping system. As its name implies, a lane-keeping system actively seeks to keep the vehicle within its current lane by applying anything from steering torque overlays to differential braking commands.
At the far end of the vehicle autonomy spectrum lie the autonomous systems. The more technical/nerdy among you may have heard or read about these during the DARPA Urban Challenge, and DARPA Grand Challenge competitions. Or, you might remember Nightrider. Autonomous systems are designed to navigate a vehicle without any human input.The common approach is to plan a path through the environment given sensory information about the location and velocity of obstacles, then track that path using a suitable controller.
ADAS system classes arranged in order of increasing vehicle autonomy
THREAT ASSESSMENT
At the heart of each of these sytems lies the need to determine, based on the current state of the vehicle, the environment, and (optionally) the driver, the level of threat the current situation poses to the vehicle.The algorithms used to make this evaluation are known as threat assessors and can be argued to be perhaps the single most important component of any active driver assistance system. Without an accurate assessment of threat, these systems can be ineffective at best (imagine a collision warning sound incessantly and unnecessarily dinging as you drive) and downright deadly at worst (as when a lane keeping controller misreads a lane marking and sends the vehicle careening off the road). Figure 1 illustrates the relationship between threat assessors and ADAS systems.
Figure 2. Venn diagram illustrating the relationship between threat assessment and various modes of driver assistance.
Given the paramount importance of an accurate threat assessment, it becomes the task of any designer to determine what exactly constitutes “threat” and how to combine various sources of threat in some meaningful way. For example, imagine a vehicle traveling down an urban road amidst other vehicles. Potentially, any vehicle up ahead could pose a threat to the host vehicle were it to slow, stop, or lose control. Similar arguments can be made for other obstacles such as pedestrians, bikers, pets, curbs, etc. How, then, does one evaluate the threat each of these obstacles poses to the vehicle? Furthermore, how does s/he combine the threat posed by various (and often very distinct) sources into a single metric upon which s/he can base the decision as to how to best assist the driver? Further complicating this task is the requirement that not only must the vehicle avoid colliding with obstacles, but it must also stay within its own lane (or whichever lane the driver chooses), remain on the road, and avoid skidding, rolling, or otherwise losing control.
Though these questions may sound philosophical, they have significant practical implications for an intelligent assistance system, which must determine whether, when, to what degree, and in exactly what manner to intervene to help the driver simultaneously avoid collisions, instability (skidding, rollover, etc.), and loss of control. In my next post, I will discuss how existing ADAS systems assess threat and seek to assist the human driver and describe how these approaches fall short of the ultimate goal of comprehensive advanced driver assistance. I will then describe the alternative: our threat assessment and semi-autonomous control system.
Electricity. In the developed world, electricity is simply interwoven into our lives to the point where most people don’t think about it unless it’s suddenly not there. Although it often gets a bad rap for when blackouts and other disruption events occur, let’s not forget that national and international electricity systems (especially the one in the US and Canada) are often called the largest “machines” in the world. Considering their size and the number of interconnected parts, electricity systems work impressively well. The fact that most people don’t think about electricity except for the moments it’s not there is arguably evidence of that. In recognition of this impressive “machine”, ‘electrification’ was named the top engineering achievement as ranked by the National Academy of Engineering’s ‘Greatest Engineering Achievements of the 20th Century’!
So, is the electrical grid broken? Arguably (which I will not do here), no.
Is there a strong desire to upgrade the system so it operates better and more efficiently? Yes.
The so-called “smart grid” is a hot topic now, with a major influx of investment coming from the 2009 American Recovery and Reinvestment Act’s Smart Grid Investment Grants. However, the grid will not change overnight. In my mind, upgrading the largest “machine” in the world will be a continuous evolutionary process.
For me, this is the connection to the Design Impact blog, and I’d like to thank Dr. James Allison for inviting me to write a guest entry about the smart grid. The smart grid will ultimately have many levels of design. How should we design the smart grid? How should we design the consumer products that will interact with the smart grid? How do we design the evolution to the smart grid while continuing to operate the grid in whatever state it is currently in? With apologies to whoever said this originally (as I have forgotten), an analogy I particularly like is that upgrading the current grid to the smart grid on the fly is effectively equivalent to changing the engine on a commercial jet while it’s flying.
Designing and managing the smart grid evolution will be a huge challenge, although not insurmountable. Ultimately, designing the underlying enabling infrastructure for the smart grid will be key. At the moment, we simply aren’t sure which technologies or systems will work best for the smart grid. To address this, I am a firm believer in experimenting and trying new technologies in demonstration projects, which is precisely the point made recently by Patricia Hoffman, DOE’s assistant secretary for electricity delivery and reliability. The Smart Grid Investment Grants are certainly a solid start at funding some experimental smart grid designs. Some ideas will work, some won’t. As these demonstration projects progress, there will be a desire to keep what works and jettison what doesn’t on the fly, meaning that the smart grid will always be in a state of transition. So, how do we design the smart grid to continuously operate under continuous change?
I return to my point earlier that the underlying enabling infrastructure will be key. One effort to help support this goal is the monumental task being spearheaded by NIST to establish communication standards for the smart grid. Among other things, smart meters, utility energy management systems, home energy management systems, and even appliances will need to be able to ‘talk’ with one another. The full spectrum of devices that will connect to the smart grid will almost certainly come from more than one manufacturer, much like a multitude of devices connects seamlessly to the Internet. Establishing communication and interoperability standards is thus critically important for innovation to flourish on the smart grid just as it has on the Internet.
Smart meters are also undoubtedly a key enabling piece of the smart grid’s evolution. Electricity usage is read off of older meters at a frequency of at most once a month, whereas these smart meters will be read on the order of a few minutes to hourly. With this more frequent feedback of electricity usage, electricity customers will have a better understanding of how much electricity they use and at what times they use it. However, smart meters are just a starting point, and as a few utilities have found out, there will be some growing pains along the way as we transition into the smart grid.
These growing pains are likely part of what was behind a recent announcement that had the smart grid world buzzing: the Maryland Public Service Commission (PSC) turned down Baltimore Gas and Electricity’s smart meter rollout proposal. Personally, I think the Maryland PSC made the right call for reasons along the lines of what Chris King discusses in an article for SmartGridNews (which is a smart grid newsletter that I recommend perusing for anyone interested in easy reading and quick introductions to the many movers and shakers in the smart grid space). It’s not that the Maryland PSC doesn’t support the smart grid. Quite the opposite, I believe. My interpretation of their reasoning is simply that ‘we like where you’re going, but we think your smart grid system design should be better.’ Designing these systems is, frankly, going to be hard. Some pieces, like smart meters, are necessary enablers of the smart grid, but there is much more to truly make the system work. There are many questions to answer as well. Among them, how will customers react in the long run to smart meters, real-time electricity information and possibly time-varying pricing? Will the new smart grid system truly operate more efficiently than the old system? Again, one of the best ways to find this out in my mind is to try out some ideas through demonstration projects, just as Patricia Hoffman suggested.
I’ll stop here for this entry and return at a later date with some thoughts on one or more of the other pieces of the smart grid. I welcome any comments, questions or suggestions of which topic or topics to discuss next.
Once again, many thanks to Dr. James Allison for providing me the opportunity to write this guest entry for his Design Impact blog. Have a great day, everyone!
On the one year anniversary of Design Impact (Earth Day 2010), I thought I would share some thoughts about how my experience as an engineer has shaped my view of the natural world. The things engineers create can be phenomenally complex, challenging and surprising their makers. We know a lot about engineered systems (they were created by people after all), but we don’t understand them completely. It may be easy to understand their constituent parts, but because of the numerous direct and indirect interactions within a system, understanding how the overall system behaves is a more demanding task. It’s difficult to conceptualize how a small change might propagate throughout a system. Engineering experience has taught us that as systems increase in complexity, the consequences of change tend to be more profound. People often get first-order effects right, but some non-intuitive outcomes are the result of a chain reaction several layers deep. For example, engineers thought they understood the behavior of the Millennium Bridge very well before opening day, but were in for a surprise:
In hindsight the interaction between the sideways bridge motion and how people walk is clear, but it eluded engineers until it was too late.
Now take a moment and consider what we know about natural systems. They are resilient, elegant, and essential to human survival. We have studied the natural world and have remarkable (but incomplete) knowledge of it. As with engineering systems, we might have reasonable component-level knowledge, but our comprehension of the intricate inter-dependencies within natural systems is truly embryonic. Lack of system-level knowledge hinders our ability to predict the full consequences of human influences. We were caught off-guard by the results of a single interaction in the Millenium Bridge system - something that we built! What then can we expect when we mess with systems that we did not create, systems with structure only partially revealed through our observation and study?
Humans have several advantages when it comes to understanding engineered systems. We made them and know how they are put together. We can consult specifications and computer models used in their design. In contrast, we don’t have access to design plans for sophisticated natural systems that have evolved and adapted over millennia. We are constantly discovering new relationships and behavior, as well as the importance of seemingly insignificant species in ecosystems. As John Muir once said, “When we try to pick out anything by itself, we find it hitched to everything else in the universe.” The intricate links between elements of the natural world are astounding and humbling, surpassing by magnitudes the complexity of mankind’s most sophisticated creations. We can understand and predict correctly the effect of some disturbances on natural systems, but the full ramifications of human impact are likely to be more extensive and deeper than we expect — far more surprising than the wobbly bridge.
Even with modern analysis tools, predicting the results of substantial changes in engineered systems is somewhere between hard and impossible. To avoid unpleasant surprises when designing especially complex systems (automotive design, for example), engineers typically put forward designs that are essentially small perturbations of previously proven systems. We are conservative and resist ambitious changes in engineered systems, yet for some reason (economic externalities?) humans are quick to risk big impacts (pollution, unsustainable resource depletion) on the natural systems we depend on. Some dismiss the notion that humans can have extensive impact, even labeling this idea as arrogant. This convenient rationalization for continued consumption growth is short-sighted and blind to history. Human disruption has caused collapse of ecosystems, even whole societies. While past collapses have been regional in scope, modern society is more populous, resource intensive, and globally interdependent than ever, enhancing our potential for impact.
In summary, we need to recognize the limits of our ability to predict the consequences of human disruption; these consequences are likely to be more profound than we expect. Our interest in the long-term health of natural resources and ecosystems provides incentive to be conservative in our consumption and impact. Our current trajectory cannot be maintained; no system can keep expanding without bumping into limits. Planning and self-imposed restraint are more pleasant options than waiting until we run up against hard constraints such as resource depletion. As the most intelligent and powerful earthly inhabitants, stewardship to preserve is ours. Over the last year Design Impact has addressed ways to leverage our intelligence to provide a high quality of life without applying unsustainable pressure on our world, and will continue to explore how we can create a brighter future for ourselves.
A major theme of Design Impact is how better engineering analysis and design can improve the sustainability of our society. One important way this can be realized is through advancing renewable energy sources, as well as improving energy efficiency. For example, advanced design techniques can help us make wind turbines and solar arrays more effective, as well as bring them online faster. In addition to addressing the resource side of the energy issue, advanced design techniques, such as design optimization, can help engineers develop transportation systems, buildings, and other engineered products that consume far less energy, while still meeting performance demands.
We have an opportunity tomorrow to learn from an impressive array of speakers at the MathWorks Virtual Energy Conference. Anyone can register (free) to watch and listen to the speakers, or to network with other participants. Many of you have probably already heard of or participated in virtual conferences, an emerging trend. If not, the basic idea is to capture many of the benefits of attending a conference in person, but via a virtual environment. You can participate from your home or office computer. I hope to see you at tomorrow’s conference!
This week is National Engineering Week, and today is Introduce a Girl to Engineering Day. There are events across the U.S. throughout this week focused on both encouraging students to consider engineering as a profession, and to help everyone deepen their understanding of what the profession of engineering is about. In Boston we had a two-day long program with design competitions, career guidance, and a career fair. What are some e-week events happening near you?
While I’m certainly an advocate of encouraging more students to consider engineering as a profession, I’m especially interested in e-week as an opportunity for the public to learn what engineering is about: what engineering has done in the past to help humanity, and the potential it has to address some of society’s most pressing present challenges. In fact, emphasis on the role of engineering in society could stimulate more interest in engineering as an attractive career choice. Senator Ted Kaufman (D-Del.), the only engineer in the Senate, explained recently that one of the road blocks in encouraging more students to pursue science and engineering careers is that they “don’t view engineering and science as the way to make a difference”, but then points out several critical issues that depend on a strong engineering workforce, including energy and economic recovery.
A clear theme throughout Design Impact articles is the positive impact engineers have on humanity. What do you see as the most important issues today that call for engineering solutions? How can we communicate best to students that a career in engineering is an opportunity to make an important difference?
An article in last month’s PRISM, a magazine published by the American Society for Engineering Education, discusses the value of a first-year engineering course that exposes freshman engineering students to what engineering is all about. Many engineering programs pack their first year with challenging prerequisite courses, such as calculus, physics, and chemistry, but sometimes neglect helping students get the big picture early on. It’s easy for a student to get lost in the labyrinth of technical topics, and lose sight of what engineering is all about.
The author of the PRISM article, Prof. Henry Petroski of Duke University, advocates including ‘engineering appreciation’ courses in the engineering curriculum, and focuses on the value these courses have to engineering students. Petroski likens engineering appreciation courses to other introductory courses offered in other disciplines that have no prerequisites, such as art or art history appreciation.
I was fortunate enough to experience two different introductory engineering classes at two separate universities. Each of these courses involved an engaging design project and competition that helped students experience the engineering design process covered in class. In the first course the project was to build a trebuchet (a weight-powered catapult) for launching golf balls. In the second class we built a device that could ride down a model roller coaster, and safely rescue an egg positioned below the roller coaster. In each class I learned something about what engineers actually do in a fun and engaging way; I began to develop my own vision for what I wanted my engineering career to be.
I believe that developing a personal vision for what engineering is (as a profession and how it impacts the world) is essential for all engineering students. This vision can help carry students through the demands of their engineering program, and help them derive more relevance from the individual topics they study. Engineering appreciation courses are certainly a valuable in this regard. I would like to take this idea to the next level. Let’s not limit these courses to engineers or prospective engineers. The art appreciation courses in Petroski’s comparison are not limited to art majors. Engineering, science, and business students all benefit from taking classes like art appreciation, which help them develop a more well-rounded understanding of the world. Why not develop an engineering appreciation class open to all students, targeted specifically for non-engineers? Obviously an engineering appreciation class benefits future engineers, but what about an even broader impact? What would it mean to our society if many college graduates had a solid understanding and appreciation for what engineering is? It could do wonders for the public perception of engineers, and perhaps even contribute to restoring U.S. economic competitiveness by inducing deeper appreciation for and stronger cultural value for technical skills and innovation.
My vision for an engineering appreciation class is one with no prerequisites that college students from all majors could take to fill a science general education requirement; students could take this instead of physics or chemistry. It could be centered around interesting applications students can relate to, things like the engineering behind sports, amusement parks, video games, music, etc., and show how basic math and science topics are relevant to engineering analysis and design. If you were a non-engineering college student, would you consider taking an engineering appreciation class in place of physics? What ideas do you have that could make an engineering appreciation course appealing to a broad range of students?
