Menu

ME team Wins award in the Clemson Covid Challenge

July 21, 2020

Delivery Drone

Delivery Drone

A team of 5 students from ME department, computer engineering and computer science have been working with Dr. Yiqiang Han on the Clemson COVID challenge competition this summer. Within only 4 weeks, the students were able to come up with a functional prototype of a complex autonomous delivery platform for this pandemic. Please see the video linked here for more details. https://www.youtube.com/watch?v=FzGmHuBm_Ec&feature=youtu.be Here is the team’s website: https://spark.adobe.com/page/HhBx6NVfi52jZ/

Team members included Yiqiang Han, Faculty Advisor, Research Assistant Professor; Rongyao ”Tony” Wang, Mechanical Engineering (ME), Graduate Student; Wenjian Hao, Mechanical Engineering (ME), Graduate Student; Alexander Krolicki, Mechanical Engineering (ME), Senior; Duncan McCain, Computer Engineering (ECE), Senior; Connor Willoughby, Computer Science (SoC), Junior; Brandon Wingard, Computer Science (SoC), Junior; Jonathan Daniel, Computer Science (SoC), Sophomore.

The Clemson COVID Challenge is a virtual research and design opportunity aimed at working on solutions to problems related to the current COVID-19 pandemic as well as possible future pandemics. Students were given 4 weeks to develop their products and present an e-poster plus a 3-minute video pitching their ideas to a wide range of audience from industry to academia (Clemson and Univ. of South Carolina). A total of 550 individuals signed up, forming closed-to 100 teams.

Team H33 presented their research on “Prototype Design for an Off-road Autonomous Delivery Platform during a Pandemic.” The team was awarded the COVID Challenge Excellence Prize made available through funding from the Clemson CECAS Dean’s Excellence fund. The award was selected based on excellent scores and feedback from 7 judges. The judges found that the project best leveraged Clemson University’s strengths over other competing projects.

Clemson Covid Challenge Team

Clemson Covid Challenge Team

The team designed an autonomous system that safely delivers packages, reduces human-to-human contact to protect customers, and strengthens current supply chain logistics for last-mile delivery. Students are working on a prototype of the last-mile delivery platform that consists of an off-road ground delivery vehicle, which also houses a delivery drone. Students are solving key technical issues relating to aerial-ground coordination, human intent prediction and robot-human interaction, motion-planning optimization for a multi-body multi-agent system, etc. The team has successfully demonstrated how a deep neural network combined with novel navigation algorithms can help robot agents navigate through paths and trails. Students were able to get the most out of their research experience by having the opportunity to perform experiments on physical prototypes compared to just working in a simulated virtual environment.

Now more than ever, people are relying on delivery services to obtain their most essential goods, and current logistical approaches for delivery are being strained. The proposed technology could lessen the burden by making the last mile drop-off of goods more time-efficient and safer for package carriers. If public testing becomes widely available, test kits could also be more efficiently distributed by utilizing our autonomous vehicles. The proposed idea would potentially boost the economy by making access to goods and services safer and more attainable for a large number of people while also ensuring that everyone is practicing safe social distancing guidelines.
 
The team designed an autonomous system that safely delivers packages, reduces human-to-human contact to protect customers, and strengthens current supply chain logistics for last-mile delivery. Students are working on a prototype of the last-mile delivery platform that consists of an off-road ground delivery vehicle, which also houses a delivery drone. Students are solving key technical issues relating to aerial-ground coordination, human intent prediction and robot-human interaction, motion-planning optimization for a multi-body multi-agent system, etc. The team has successfully demonstrated how a deep neural network combined with novel navigation algorithms can help robot agents navigate through paths and trails. Students were able to get the most out of their research experience by having the opportunity to perform experiments on physical prototypes compared to just working in a simulated virtual environment.