Vehicular crowd sensing (VCS), emerging as a new paradigm within mobile crowd sensing, leverages vehicles as the participator, which can obtain broader sensing coverage and higher sensing flexibility. Previous works ignored the strong impact of environmental factors on workers' travel costs, as well as improper gains from speculative behavior (i.e. workers detour or delay to get more compensation), resulting in unfair income of workers. Moreover, these works focused solely on sensing tasks within specific domains, lacking generalization ability. Therefore, our work is dedicated to providing a fair and universal VCS platform, which is called Environment-cost-based Heterogeneous Task Allocation (EHTA) framework. Our work differs from previous works in the following aspects: 1) We introduce the Environment Cost (EC) based on the investigation of traffic conditions to accurately quantify workers' efforts, and propose a straightforward yet efficacious detection methods to identify speculative behavior of malicious workers, both of which could guarantee the fairness in workers' income. 2) We design a spatial-temporal fair incentive mechanism based on monetary reward to ensure the fair execution of tasks in both space and time dimensions. 3) We summarize the characteristics of three kinds of sensing tasks and propose a universal task allocation algorithm to assign multiple types of tasks simultaneously. The effectiveness of our framework was validated by simulations, which are conducted on a data set comprising 13,000 taxi trajectories from Shanghai in April 2015. We compared our framework against four baseline algorithms, and the results shows that EHTA framework outperforms in terms of task expenditure, task utility and fairness.