The continuously variable transmission (CVT) is one of prevailing automotive transmissions. Nowadays, the most popular design is Van Doorne’s CVT with single metal pushing V-belt. However, it is only applicable to low-power passenger cars because of its low torque capacity. To overcome this limitation of traditional single-belt Van Doorne’s CVT (SBVCVT), this research proposes a novel dual-belt Van Doorne’s continuously variable transmission (DBVCVT) system which can be applicable to heavy-duty vehicles. As the first attempt, an analytical model for both torque capacity and power efficiency based on effective radii instead of traditional working radii for the single-belt CVT is developed for this novel CVT system. Moreover, this research originally fabricates a prototype DBVCVT and integrates it to a test rig for testing. The comparison of simulation and experimental results validates this newly developed model and some characteristics of the DBVCVT in terms of torque capacity, speed ratio response and transmission efficiency. Evaluation result reveals that the new analytical model based on effective radii is more accurate than that based on working radii. Experimental and simulation results show that the proposed DBVCVT can greatly improve the torque capacity as compared with the single-belt CVT. A case study based on the analytical model shows the proposed DBVCVT has a potential to be used on heavy-duty vehicles. In addition, this research originally solves the coupled slip and speed ratio control problem for the DBVCVT by using an intelligent multi-objective fuzzy controller. The controller aims to concurrently control the clamping forces of both primary and secondary pulleys in order to improve the transmission efficiency, achieve the accurate speed ratio and avoid the belt slip under different engine loads and vehicle speeds. With the aids of a flexible objective function, analytical model and fuzzy logic, a Pareto rule base for the fuzzy controller is also developed for the multi-objective DBVCVT control. Besides, a fuzzy-proportional-integral-derivative (fuzzy-PID) controller with mapping feedforward for the slip and speed ratio decoupling control is developed for comparison. Experimental results show that compared with the decoupling control, the proposed multi-objective controller for the coupled slip and speed ratio control of the DBVCVT is effective and performs well under different user-defined weights.