Incremental Sheet Forming (ISF) is a new-emerging sheet forming process that promises high flexibility and formability because it does not need any dedicated dies or tooling. This makes ISF well suited for small scale and customised production. ISF uses a simple tool with a smooth end to deform sheets along a toolpath typically generated from CAM software. In ISF, the part is fabricated incrementally by orderly accumulation of plastic deformations localised around the ball end of the tool. Without using dedicated tooling, ISF can quickly adapt to new and modified product shapes via toolpath alteration. Hence, ISF is a promising sheet forming process for quick and low-cost production for small batch manufacturing. In spite of these outstanding characteristics, ISF still suffers from some drawbacks including long processing time and low geometric accuracy. The latter one is the major cause of low take-up of ISF in the forming industry. Therefore, the work presented in this thesis is mainly focused on the improvement of geometric accuracy of the formed parts via inprocess toolpath correction using feedback control. In this thesis, Model Predictive Control (MPC), an advanced model-based control technology, was adopted to develop feedback control strategies for ISF toolpath control and correction to improve geometric accuracy.The first research aspect of this thesis is focused on the development of a simple MPC algorithm to optimise the step depth, a critical toolpath parameter defined by the user in the toolpath generation.Firstly, a parameter investigation was performed to experimentally study the influences of the step depth on geometric accuracy, surface finish, and thickness distribution. Two sets of experimental tests with varying step depth values were conducted in the typical Single Point Incremental Forming (SPIF) process that is without toolpath control. The results showed that a smaller step depth led to better geometric accuracy and part surface quality in the ISF process, while the material formability decreased with the step depth value. Too small step depth values would lead to material failure. The parameter investigation work provided significant fundamental information for the development of feedback control strategies for toolpath correction. Then, a simple MPC algorithm was developed for in-process toolpath control/correction in SPIF. During the forming process using the simple MPC algorithm, the step depth of the contour toolpath was optimised at each forming step by solving a receding optimal problem based on the measured shape feedback during the forming process. The developed algorithm was experimentally verified in two case studies to form two different shapes.Results show that the geometric accuracy in ISF with feedback control has been greatly improved (from ±3 mm to ±0.3 mm) at the bottom area of the formed parts compared with a standard ISF approach without control. Improved geometric accuracy has been achieved on the wall of the parts as well, but the errors in the wall areas are...