On The Quantum Algorithm For Collaborative Navigation And Quantum Gates Discrimination Under The Perspective Of Intelligent Agent

The continuous development of quantum physics has fused with other disciplines to derive many new research fields.Applying quantum information to the agent model will make the agent to have certain characteristics that the traditional agent does not have,which has important research significance.To this end,research are conducted from three different level in this thesis,which is described as follow.（1）We propose a paradigm of hybrid agent in which quantum information and classi-cal information coexist.The behavior of an agent mainly consists of perception,decision-making and action.Therefore,the paradigm of the agent can be defined as a five-tuple.By redefining the five-tuple,this thesis proposes a new agent behavior paradigm—the hybrid agent paradigm.By comparing the agent paradigm with the previous two agent paradigm,It is found that the hybrid agent paradigm can better describe the combination of quantum information and agent.（2）We study the quantum-classical hybrid collaborative navigation problem under the paradigm of hybrid agent.Based on the classical Kalman filter algorithm,two quantum-classical hybrid algorithms are designed,namely LE-QCKF algorithm under low entan-glement and HE-QCKF algorithm under high entanglement.We assume that each agent measures the distance of other agents in collaborative navigation.So the dimension of innovation will be much larger than the dimension of the system state when the scale of the multi-agent system n is large enough.The following conclusions can be drawn through theoretical analysis.If the precisionεsatisfies 1/ε=O（n）,the two quantum-classical hy-brid collaborative navigation algorithms reduce the time complexity from O（n⁷）to O（n⁴）;and if the accuracy satisfies 1/ε=O(n^2√n),the two quantum-classical hybrid collabo-rative navigation algorithms have an acceleration effect in time complexity.That is to say,when the scale of the multi-agent system is large enough,the superiority of the quantum algorithm can be exerted within a certain accuracy range.In addition,the HE-QCKF al-gorithm achieves an exponential reduction compared to the LE-QCKF algorithm in terms of space complexity by using quantum entanglement.Finally,simulation experiments of small scale agent system are carried out in the qiskit quantum simulation platform devel-oped by IBM to preliminarily verify the feasibility of the two quantum-classical hybrid algorithms.（3）we carry on several problems of multiple quantum gates’discrimination under the sequential structure.Multiple quantum gates’discrimination is studied from the per-spective of a multi-agent system under the hybrid agent paradigm,which provides new ideas about multiple quantum gates’discrimination and even more quantum information problems.Then two different schemes for the discrimination of multiple quantum gates are given:two-by-two discrimination scheme and one-time discrimination scheme.The comparison of the two schemes shows that the quantum gates in the later have much less using frequency than the prior.In essence,the discrimination efficiency is actually im-proved by the quantum entanglement.In conclusion,quantum entanglement can improve the performance of computing,as well as the efficiency of solving quantum information problems.