US 11,887,485 B2
Control method and system for collaborative interception by multiple unmanned surface vessels
Huayan Pu, Shanghai (CN); Yuan Liu, Shanghai (CN); Jun Luo, Chongqing (CN); Zhijiang Xie, Chongqing (CN); Jiajia Xie, Shanghai (CN); Xiaomao Li, Shanghai (CN); Zhou Su, Shanghai (CN); Yan Peng, Shanghai (CN); Hengyu Li, Shanghai (CN); and Shaorong Xie, Shanghai (CN)
Assigned to Shanghai University, Shanghai (CN); and Chongqing University, Chongqing (CN)
Filed by Shanghai University, Shanghai (CN); and Chongqing University, Chongqing (CN)
Filed on Sep. 7, 2021, as Appl. No. 17/467,916.
Claims priority of application No. 202110012035.7 (CN), filed on Jan. 6, 2021.
Prior Publication US 2022/0215758 A1, Jul. 7, 2022
Int. Cl. B63B 49/00 (2006.01); G08G 3/00 (2006.01); B63B 79/15 (2020.01); B63B 79/40 (2020.01); B63B 79/20 (2020.01); G05B 13/02 (2006.01); G05D 1/02 (2020.01)
CPC G08G 3/00 (2013.01) [B63B 49/00 (2013.01); B63B 79/15 (2020.01); B63B 79/20 (2020.01); B63B 79/40 (2020.01); G05B 13/0265 (2013.01); G05D 1/0206 (2013.01)] 10 Claims
OG exemplary drawing
 
1. A control method for collaborative interception by a plurality of unmanned surface vessels, which is performed in a computer, comprising:
obtaining task environment information of each unmanned surface vessel in an unmanned surface vessel group at a current moment; wherein the task environment information comprises position information of the unmanned surface vessel, velocity information of the unmanned surface vessel, relative position information between the unmanned surface vessel and an intruding target, and relative velocity information between the unmanned surface vessel and the intruding target; each unmanned surface vessel corresponds to an intruding target interception policy output model, and the intruding target interception policy output model is obtained by training a multi-agent deep deterministic policy gradient network structure based on a training sample;
estimating interception point information of the intruding target at the current moment by using a Kalman filter according to the task environment information of all unmanned surface vessels at the current moment; wherein the interception point information comprises a predicted position of the intruding target and a predicted velocity of the intruding target;
determining process state information of each unmanned surface vessel at the current moment; wherein the process state information comprises the position information of the unmanned surface vessel, the velocity information of the unmanned surface vessel and the interception point information of the intruding target; and
inputting the process state information of each unmanned surface vessel at the current moment into a corresponding intruding target interception policy output model respectively to obtain an execution action of each unmanned surface vessel at a next moment, and sending the execution action to an execution structure of a corresponding unmanned surface vessel to intercept the intruding target.