Search
On July 18, news emerged that the 2026 Global Artificial Intelligence Conference took place from July 17 to 20. During the event, Qianjue Robotics presented its groundbreaking VTLA (Vision-Touch-Language-Action) embodied tactile model, alongside a visual-tactile multimodal dataset, core data acquisition tools, and a tactile sensor hardware array.
As explained by Ma Daolin, the visionary founder of Qianjue Robotics, the VTLA model marks a significant departure from conventional single-vision VLA models. It ingeniously fuses visual and tactile bimodal perception data, empowering robots with both visual recognition and tactile sensing prowess. This unique integration enables robots to discern physical attributes of objects that elude single-vision devices, thereby substantially broadening the scope of recognition and refining detection precision.
-
C114 Communication Network -
Communication Home
