Chongqing - A student team from Chongqing University of Technology (CQUT) has won the national first prize in the Bionic Legged Robot Challenge at the 25th National College Student Robot Contest (ROBOCON), held in Jiangsu Province. The achievement marks the best-ever result for a Chongqing university in this competition category.
The Shiji Robot Team from Chongqing University of Technology claimed the national first prize in the Bionic Legged Robot Challenge at the 25th ROBOCON, recently held in Jiangsu. (Photo/Chongqing University of Technology)
The ROBOCON contest, which began in 2002, is widely regarded as one of China's most technically demanding and influential robotics competitions for university students. The champion team earns the right to represent China at the ABU Robocon, the Asia-Pacific regional tournament organized by the Asia-Pacific Broadcasting Union.
This year's event, hosted at Nanjing University of Science and Technology's Jiangyin campus, attracted nearly 5,000 participants from 297 teams representing 133 universities and colleges nationwide. The competition featured three categories: a "Martial Arts Discovery" robot challenge, a bionic legged robot challenge, and a robot volleyball challenge.
Ten months of development
The CQUT team, named "Shiji Robot Team," consists entirely of undergraduate students from the university's School of Mechanical Engineering. The 23 core members represent multiple majors, including Mechanical Design, Manufacturing and Automation, Mechatronic Engineering, Measurement and Control Technology and Instruments, and Intelligent Manufacturing Engineering. The team spent ten months developing their award-winning bionic legged robot.
Team members in action during the competition. (Photo/Chongqing University of Technology)
Innovative wheel-leg integrated design
"The most ingenious feature of our robot is its series wheel-leg integrated design," said Wu Lingran, captain of the Shiji Robot Team. The robot's joints are powered by RobStride 02 motors, with each leg integrating both legged motion joints and an independent drive wheel module. This design enables seamless switching between high-speed wheeled locomotion on flat terrain and flexible legged gaits for obstacle negotiation—without requiring additional switching mechanisms.
Team members are testing the robotic products. (Photo/Chongqing University of Technology)
On flat surfaces, the robot operates in wheeled mode, achieving over 60% greater travel efficiency than pure legged robots. When encountering obstacles such as cobblestones, wood chips, steps, or slopes, it automatically transitions to legged gaits, using its multi-degree-of-freedom legs to adapt to terrain and clear hurdles—perfectly suited to the competition's mixed-terrain format.
Reinforcement learning ensures reliability
To enhance motion intelligence and stability, the team adopted a reinforcement learning-based simulation-to-reality training approach, which ensured zero operational faults throughout the competition, Wu added.
Team members are testing the robotic products. (Photo/Chongqing University of Technology)
In addition to securing first prize in the obstacle race, the CQUT team earned three national third prizes—two in the main competition track and one in the Bionic Legged Robot Task Race.