Millimeter-wave (mmWave) radar has become an important sensing technology for intelligent transportation, urban traffic management, perimeter security, and industrial automation because of its robustness under low illumination, rain, fog, dust, and other adverse conditions. However, most existing studies focus on single-domain optimization, while limited attention has been paid to how radar-vision fusion frameworks should be adapted under heterogeneous deployment constraints such as tunnel multipath, dense urban occlusions, security false alarms, and industrial latency requirements. This study proposes a domain-adaptive radar-vision fusion framework integrating spatiotemporal calibration, adaptive constant false alarm rate (CFAR) processing, and joint probabilistic data association (JPDA)-based multi-target tracking. The framework is validated through field deployments on the Nanning and Qinzhou sections of the G75 Lanzhou–Haikou (Lanhai) Expressway and the Fenghuang No. 1 and No. 2 tunnels on the Laibin–Du’an section of the Hezhou–Bama (Heba) Expressway in Guangxi, China, and further analyzed through structured cross-domain extension application and analysis involving urban traffic, perimeter security, and industrial sensing. The results show that the proposed framework achieves robust vehicle detection and trajectory continuity under low-light, exhaust-smoke, and multipath conditions, while domain-specific hardware configurations and algorithmic thresholds are necessary to satisfy different sensing ranges, clutter characteristics, and latency constraints. The study identifies key limitations of current 3D multiple-input multiple-output (MIMO) radar deployments. These findings provide practical guidance for the transition toward 4D imaging radar and edge-collaborative perception infrastructure.

The supply of high-quality talent plays a decisive role in the sustainable development of the highway industry. To accurately capture trends in AI–highway engineering interdisciplinary talent and to inform talent cultivation and policy formulation in higher education, this study analyzes data from Fujian Province spanning 2019–2024, including the total number of highway engineering professionals, total industry investment, talent attrition, and the supply of college graduates. A Bayesian regression model is employed for predictive analysis. The results indicate that parameter estimates are consistent with prior assumptions, the sampling process is stable, inter-chain convergence is satisfactory, and parameter estimates are reliable. The posterior distributions are approximately normal, and the Markov chain Monte Carlo (MCMC) trajectories exhibit random behavior without discernible trends, indicating good chain mixing and robust posterior estimation. Overall, demand for AI interdisciplinary talent in the highway engineering sector of Fujian Province shows a significant upward trend. Under the baseline scenario, the total industry workforce is projected to reach 5,351 by 2029, including 1,605 AI interdisciplinary professionals, representing a cumulative increase of 687 over five years. In the optimistic scenario, driven by increased investment, reduced drain, and growth in graduate supply, the workforce is expected to expand to 6,508 by 2029, with 1,952 AI interdisciplinary professionals, a cumulative increase of 999 over five years. Even in the pessimistic scenario, despite adverse conditions such as reduced investment and higher turnover leading to a workforce decline to 4,743 by 2029, the number of AI interdisciplinary professionals is projected to continue growing, reaching 1,423, with a cumulative increase of 522 over five years. These results demonstrate that the Bayesian regression model effectively quantifies the dynamic effects of total highway industry investment, brain drain, and graduate supply on workforce scale, and that the forecasts are consistent with practical industry constraints.