Ditch the snow shovels, the robot snow ploughs are in town
Communities up and down America's east coast states have been battling heavy snow fall in recent weeks as winter storm Juno did her worst. But whilst the snow shovels were out in force, little did they know that further west in Minnesota a battle of a different kind was being fought that might just help them out in future storms.
The Fifth Annual Autonomous Snowplow Competition took place on 24-25 January 2015 in Saint Paul, Minnesota. Founded by the Institute of Navigation (ION), this year's competition saw eight teams from six universities facing the challenge to design, build and operate a fully autonomous snow plough.
This year's winners, Team Zenith 2.0, representing the University of Michigan-Dearborn, took home the prestigious Golden Snow Globe Award together with $7,000 to be used to develop their design further.
Students are encouraged to use state-of-the-art navigation and control technologies that will rapidly, accurately and safely clear a designated path of snow. The competition has a number of objectives such as encouraging the use of maths and science to solve every day real-life problems, as well as promoting the value of these skills to students and the general public alike.
"The most important objective is to motivate and excite undergraduate engineering students about guidance, navigation and control systems by building snow ploughs," explains Dr Vibhor L Bageshwar, a senior scientist in advanced technology at Minneapolis-based Honeywell Aerospace, one of the competition's organisers.
The competition is held in conjunction with the Saint Paul Winter Carnival. Now in its 129th year, the carnival draws around 150,000 visitors, providing ample scope for imparting the engineering message of the competition to onlookers.
"Undergraduate engineering students tend to do a lot of homework, spend a lot of time in the library, and not show the general public the great designs they work on. They find it a great and rewarding experience when their plough operates in front of cheering crowds and they begin to understand and gain appreciation that their hard work will one day lead to everyday products that everyone will use," Dr Bageshwar adds.
Let it snow
Although other parts of America have received more than their fair share of snow recently, the snow and weather conditions can be variable in Minnesota with no guarantee of sufficient snow for the Autonomous Snowplow Competition. For this reason, snow is brought in from a local ice rink for the duration of the competition, so that all teams get enough of the fresh stuff to test out their snow plough designs.
Which is just as well, as this year's competition saw a new rule instated for teams to negotiate.
"This year's key new rule modification was that snow had to be ploughed to specific areas of the competition snowfield to mimic snow removal of actual sidewalks and driveways. The other key rules of the competition are that snow plough vehicles must start and stop from a garage, avoid posts in the middle of the competition snow paths, remain within a boundary surrounding the competition snow paths, and plough the competition snow paths within 20 minutes," says fellow competition organiser Dr Suneel I Sheikh, CEO and chief research scientist at Aster Labs.
As if that's not tough enough, the teams are also put through their presentation paces and report-writing skills, with a score of their overall performance in all areas defining their final places on the scoreboard.
"The Autonomous Snowplow Competition is set up to mirror the requirements of a senior undergraduate engineering capstone design course and industry practices," says Dr Sheikh. "The competition teams are asked to give two presentations to mirror industry design practices. They also submit final reports describing their snow plough vehicle design, their strategy for ploughing the competition courses, and their thoughts on the commercialisation and implementation of their snow plough vehicles."
The course itself is no cakewalk. The snow plough vehicles weigh about 400-500 lbs and must autonomously plough snow from two competition snowfield paths: the single 'I' and the triple 'I'. To plough the snowfields, teams must navigate through the paths and their surroundings, using a variety of different sensors and approaches to navigate around the paths such as GPS, inertial sensors, wheel encoders, magnetic sensors, cameras, and LIDARs (remote sensing technology).
The technical challenges might be immense and the stakes are certainly high, but the event is characterised by an atmosphere of camaraderie and sportsmanship with the cold Saint Paul weather fostering a feeling of goodwill between teams as each openly cheers for each others successes and commiserates during struggles. As winning Team Zenith 2.0 advisor Dr Narasimhamurthi 'Nattu' Natarajan describes, "The competition has great walk-up crowds, who were always eager to ask questions and learn more about the autonomous snow ploughs. What is also cool about the competition is the incredible sportsmanship between the teams. This was a great learning experience for each team, and each was happy to share their secrets to help others experience different methods."
One of the goals of the competition is for a team to make a snow plough prototype that is not only autonomous, but also commercially viable and this was one of the criteria that team Zenith 2.0 took very seriously during the design process.
"When Jason designed his chassis, it cost less than $150 for all of the metal used (not including the plough)," explains Zenith 2.0 team member Benjamin Pollatz, a robotics engineering and electrical engineering undergraduate. "While other teams may use a $30,000 GPS system or a $3,000 laser sensor to navigate a snowfield, Zenith 2.0 uses only two sensors: a computer webcam (retail cost around $50), and an ultrasonic sensor (some available for as little as $1). Our system will work with any brightly coloured markers that can be placed on either side of the field. As long as they are symmetrically placed on either side at some approximate offset from the sides, the robot will navigate and clear the snow without any explicit calibration. What sets us apart from the other teams is the extreme simplicity of navigation."
"Commercialisation is a very realistic proposition," Nattu continues. "Right now, the base cost is $4,000 based on retail price for the components used. The owner will operate the robot from the comfort of the living room using a joystick controller and the robot will provide navigational assistance by ensuring the robot travels in a straight line, avoids obstacles or going off the driveway."
So, next time there's a severe weather warning it won't be the droves of cold homeowners we see toiling away with their snow shovels, but an army of autonomous robot snow ploughs, efficiently ploughing the snow away! It's the future...
The entry deadline for next year's competition is September 2015. For further information such as rules, sponsorship details and design requirements visit www.autosnowplow.com. Teams can apply for travel grants to Saint Paul, Minnesota, USA, through the ION starting in September 2015.