| Military training is an important way for military personnel to improve their combat abilities and military literacy,and it is of great significance to the enhancement of national military strength and the construction of national defense forces.However,traditional mil-itary exercises are often restricted by training funds,personnel,material supply,and venue limitations,and it is often difficult to achieve ideal training effects even with a significant amount of resources.By introducing mixed reality technology,the advantages of mixed reality devices can be utilized in some aspects of military training,allowing trainees to have a realistic visual and interactive experience while reducing material costs and im-proving training efficiency.However,due to the limitations of mixed reality hardware platforms,introducing mixed reality technology into the application scenarios of mili-tary training often encounters problems such as mixed space position mapping deviation,decreased model visibility in outdoor strong light environments,and inability to interact across devices.To address these issues,this paper has done the following work:Firstly,this paper proposes a real-time carrier phase difference technology-based method for real-time correction of mixed space positioning.Since existing mixed real-ity devices mainly rely on inertial and visual positioning technology to perceive device movement and location,their positioning accuracy cannot meet the needs of large-scale outdoor use.Therefore,this paper innovatively introduces differential positioning tech-nology into mixed reality,deploys positioning base stations on mixed reality devices to collect real-time device position and direction data,and processes this data to achieve real-time 3D transformation of the virtual space projected in the device,ensuring the stability of the virtual-real space position mapping.Experimental verification shows that by using differential positioning technology,stable and low-error space position mapping can be ensured even in large-scale outdoor movement.Secondly,this paper proposes a method for constructing mixed reality virtual scenes using real data.By introducing environmental feature data such as digital elevation infor-mation and ambient lighting information to model the virtual terrain in the mixed reality battlefield and optimize the scene construction process,the virtual scene can be more closely aligned with the constantly changing usage scenarios in reality,improving the user experience of the mixed reality virtual battlefield.Experimental results show that after optimizing the virtual space with geographical environmental feature data,the gen-erated virtual terrain matches the real-world environment,and with the post-processing of lighting perception,the virtual model has higher clarity.Finally,this paper developed a prototype of an online mixed reality battlefield train-ing system.With the support of the above technologies,this paper overcomes problems such as space positioning,outdoor lighting interference,and cross-device interaction,and enables the system to have functions such as multi-device synchronous participation in training,multi-machine scene space synchronization,and long-distance information in-teraction through network and database technologies,thus meeting the needs of mixed reality battlefield training.The system was tested and the results show that all functions of the system can run normally in the application scenario and have high performance and reliability. |