Cities use pavement management systems to ensure their pavement networks are kept in serviceable condition in a cost effective way. These systems rely on condition surveys that provide a numerical condition rating for every street. Conventional condition surveys require a rigorous in-field inspection by trained experts, which is time consuming, unsafe for inspectors, subjective, and causes traffic interruptions. This poster introduces a vehicle-mounted pavement inspection system that aims to overcome these limitations. In order to handle the complex nature of pavement condition and the challenges of mobile sensing, the system is comprised of a multi-modal sensing array that includes accelerometers, microphones, a camera, a radar array, a dynamic tire pressure sensor, and a laser height sensor. Data fusion and machine learning concepts are used to combine the results from all sensors into a singular rating that is comparable to conventional condition surveys. Contributions of this poster include an in-field application of the system in Brockton, Massachusetts. A comparison of these field results to the true pavement condition (known from an existing condition survey) demonstrates the system’s high potential.