Consumer-grade micro-electro-mechanical-systems (MEMS) inertial sensors are small, lightweight and low-cost in their raw form. A set of three accelerometers and three gyroscopes costs less than £100. A calibrated inertial measurement unit (IMU) containing the same sensors gives much better performance, but costs over £1000. This project is therefore investigating methods of getting better performance out of the raw sensors, focusing on two main themes. The first is intelligent sensor arrays, containing multiple inertial sensor triads. This technique could produce better performance than simple averaging by exploiting properties of MEMS sensors. These include common-mode errors of different sensors of the same design, the different characteristics of in-plane and out-of-plane sensors, and the complementary properties of sensors with different measurement ranges. The second theme is a user-conducted calibration scheme as part of a multi-sensor integrated navigation system. This would give most of the performance benefit of the expensive laboratory calibration at a much lower cost, by supplying the user with an un-calibrated set of sensors but including software that would calibrate them using a set of simple maneuvers. Project duration: 2011-2015; Sponsor: EPSRC and BAE Systems, Researcher: Henry Martin.