Research On Multi-sensor Management Method Based On Joint Risk Assessment

The goal of multi-sensor management is to plan limited multi-sensor resources according to different optimization standards to meet the information requirements of command and control,decision-making and other applications.Due to the multi-sensor detection performance,tracking accuracy and risk limits,to better plan multi-sensor resources for target tracking,in the process of multi-sensor collaborative detection target,effective reduction of the threatened risk and better assurance of multi-sensor safety is of great importance for multi-sensor management research.Taking convex optimization,optimal estimation and information fusion as technical means and guided by joint risk control,this paper focuses on the multi-sensor management algorithm when multiple risks change in complex scenes,so as to reasonably allocate limited sensor resources and effectively improve the risk assessment accuracy of target tracking process.The main research contents of this paper are as follows:Aiming at the problems of high radiation risk,large tracking error,poor tracking accuracy and high probability of being hit by multi-sensor detectors,a multi-target and multi-sensor management algorithm based on one-way joint risk assessment is proposed.Firstly,the sensor tracking error is evaluated by using the Posterior Cramer Rao Lower Bound(PCRLB),and the joint risk model is constructed by synthesizing the sensor radiation risk.Secondly,the weights of sensor tracking error and radiation risk are adjusted adaptively,and the multi-sensor management optimization problem is constructed with variable weighting.The optimization problem is solved by convex optimization technology to obtain the multi-sensor scheduling scheme.Finally,the improved consistent fusion algorithm is used to fuse multi-sensor target estimation that pursues the same target,and the fusion estimation is fed back to each multi-sensor,so as to more accurately predict the target state and manage the multi-sensor resources at the next time.The simulation results show that this method is feasible,which can reduce the radiation risk of multi-sensor and improve the tracking accuracy,so as to realize the accurate and stable tracking of multi-target.Since the unilateral risk of the sensor is not comprehensive for the joint risk assessment in the battlefield environment,it ignores the impact of the two sides’ risk on the target tracking under the attack and defense system,and a single target multi-sensor management algorithm based on two-way joint risk assessment is proposed.Firstly,the multi-step prediction of target state and the expert experience of sensor management are combined to construct the target threat risk,and joint risk is the combination of target loss risk and target threat risk.Secondly,the weights of two risks are adjusted by using the distance between the target and the sensor,and the sensor management optimization problem is constructed for minimizing the joint risk of multi-step prediction.Finally,the non-convex optimization problem is relaxed,and the multi-sensor scheduling scheme is obtained through discrete and legal processing to improve the solution efficiency of the optimization problem.The simulation results show that this method can better evaluate the sensor risk,and achieve better target tracking with the least multi-sensor resources under the minimum risk.In battlefield risk assessment,aiming at the decline of risk assessment accuracy caused by the newborn and death of tracking targets,and the location exposure caused by ignoring sensor radiation,a multi-target sensor management algorithm based on joint risk assessment is proposed.Firstly,the joint risk model is constructed by integrating the identification risk of new targets and the threat risk of existing targets.Secondly,the sensor radiation risk is constrained to improve the concealment of sensor detection,and then the sensor management optimization problem is constructed.Finally,the optimization problem is processed by convex optimization relaxation method,and then the optimized sensor scheduling scheme is obtained.Simulation results show that the proposed method can effectively control the joint risk and reduce the impact of the change of target number on sensor measurement.Aiming at the multi-sensor management algorithm is difficult to carry out multi scene simulation verification,this paper designs a simulation platform for multi-sensor management algorithm.Firstly,the requirements of multi-sensor management algorithm verification are analyzed.According to the requirements analysis,multiple functional modules such as data preprocessing,risk assessment,sensor management and data display are designed.Secondly,it introduces the main technical implementation of the simulation platform and its overall design process.Finally,the simulation verification platform is realized by using Visual C++ programming language,and the platform and its functional modules are tested and verified.