Connected vehicles need to be aware of their surrounding environments. This is impossible without being dependent on many sensory inputs. Sensor data is continually collected and analysed, in real-time in order to perform time-critical and delay-sensitive actions. There are two major challenges 1) limited computational resources (processing power and memory) on cars, 2) transfer of large sensory data to the cloud may is not feasible. In this project, we aim at building scheduling and task offloading techniques to share the computation tasks between the vehicles and cloud considering the computational resources of the vehicles, Quality of Service (QoS) requirements of the applications (e.g., required response time), and available bandwidth.