Event
DSGN Dual-Stream Gated Network for Multi-Robot Surveillance and Encirclement
1 reports · 1 sources
Heng Zhang, Ziyang Zuo, Guoxiang Zhao, and Xiaoqiang Ren propose a dual-stream gated network, DSGN, for simultaneous multi-robot surveillance and encirclement tasks. The method splits surveillance and encirclement into two independent representation streams and uses a context-aware gating mechanism to coordinate the two, so as to alleviate the difficulty of task switching as the robot network scales up, and introduces probabilistic prior regularization of the gating process to reduce ambiguous intermediate decisions. The authors state that, under constant-density cross-scale settings, DSGN maintains task effectiveness and scalability; ablation and staged experiments show that dynamic gating and probabilistic priors support task arbitration and mode switching. The paper has been accepted by RA-L for October 2026, and the code has been made public.
Report timeline
- DSGN: A Dual-Stream Gated Network for Scalable Multi-Robot Surveillance and Containment
The authors propose the Dual-Stream Gated Network (DSGN), which separates surveillance and containment into independent representation streams and coordinates them through a context-aware gating mechanism to alleviate task-switching difficulties in large-scale robot networks. The method also introduces a probabilistic prior to regularize the gating process, reducing ambiguous intermediate decisions. Under constant-density cross-scale settings, DSGN maintains task effectiveness and scalability; ablation and staged experiments show that dynamic gating and the probabilistic prior support task arbitration and mode switching. The paper has been accepted to RA-L October 2026, and the code is publicly available.
arXiv cs.RO ·