Cooperative Human-Robot Swarm Computation

IEEE Computer Society Team
Published 05/29/2023
Share this on:

Recipient of the Emerging Tech grant - Cooperative Human-Robot Swarm ComputationAccording to the United Nations, 68% of the world population is projected to live in urban areas by 2050. This will bring environmental, health, and social challenges as the population of cities swells. Communities will need to evaluate a range of technologies to address these and other urban challenges.

A program funded by the IEEE Computer Society’s Emerging Technology Fund, the Cooperative Human-Robot Swarm Computation project, was established in 2022 to develop frameworks for designing and operating cooperative human-robot swarm systems to address urban challenges. The project has constructed responses to two scenarios: firemen and fire-fighting robots in fires in buildings and cooperative truck-drone delivery in restricted traffic zones.


Firemen and Fire-Fighting Robots

China faces about 350,000 fires a year, with nearly 3,000 casualties and an economic loss of up to 4.5 billion yuan. Rescuing people trapped in fires is a critical but difficult task; buildings have varying degrees of size and wiring complexities. When a fire breaks out, and firemen enter a building, they may not know its internal structure, and the path of a fire can be extremely changeable, leading to difficulties in finding the best rescue path. As a result, rescue opportunities are missed, and lives are threatened.

Firefighting robots can increase the chance of rescues while reducing the casualties of firemen and trapped people. They can climb stairs, withstand temperatures of up to 700 degrees, and spray thousands of liters of water. Although robots cannot rescue people from a fire—especially those with limited mobility—they can make a path through flames that firemen cannot.

In their paper, “Ant Colony Optimization for Collaborative Fireman-and-Firefighting-Robot Rescue in Building Fires” (pending publication), Xiao Yang, Wei-Guo Sheng, and Yu-Jun Zheng proposed several methods, including dynamic programming (DP), ant colony optimization (ACO), and a hybrid DP-ACO algorithm to optimize the paths of firemen and firefighting robots.

The team used five underlying maps (two in urban areas and three in rural areas), and eight underlying buildings with different scales and topological structures. Each had different rescue plans and injury conditions. Rescue solutions for paths of rescue workers and robots were evaluated by average rescue time.

“This is a cutting-edge topic, which could significantly improve the efficiency and success rate of rescuing victims in disastrous fires. I really hope and believe that my study can be helpful to save lives,” said Xiao Yang, Master student at Hangzhou Normal University.


Cooperative Truck-Drone Delivery in Restricted Traffic Zones

Drones offer the potential to reduce traffic congestion by reducing the number of vehicles on the road and improving air quality through fewer idling trucks during pickups or deliveries. Many cities have implemented strict measures to restrict truck access, which can cause difficulties for local businesses. Yet drones, with their high speed and flexibility, can be used in combination with trucks for efficient delivery in restricted traffic zones.

During the COVID-19 pandemic, drones were also used for contactless delivery, especially for urgently needed medical supplies. For instance, cargo drones were used to transport medical materials to Chinese cities during the 2020 shutdowns. Many food service companies also employed drones to deliver food to customers without human-to-human contact.

In their paper, “Cooperative Truck–Drone Delivery Path Optimization under Urban Traffic Restriction,” authors Ying-Ying Weng, Rong-YuWu, and Yu-Jun Zheng propose a hybrid evolutionary algorithm and convex optimization to cooperatively improve the outer path of the truck and the inner path of the drone. A truck carrying cargo travels along the outer boundary of the restricted traffic zone to send and receive a drone, and the drone is responsible for delivering the cargo to customers. The objective is to minimize the completion time of all delivery tasks.

“I have learned a lot of valuable things: How to mathematically formulate the problem of cooperative delivery, develop efficient and smart algorithms, and write my scientific paper clearly and accurately,” commented Yingying Weng, Master student, Hangzhou Normal University.

For more information on the studies, contact Learn more about the Emerging Technology Fund.