Virtual Reality Training System For Arm Amputees Rehabilitation

The users of arm prosthesis in the case of an arm amputation are not keen on using the EMG solution to control it that recognizes the muscles signals from the residual limb.The reason is that this input is not intuitive at first,and needs a proper training to be able to effectively control the prosthetic hand in the desired way.However,users are reluctant to perform this training,time consuming,not entertaining and tiring due to the weight of the prosthesis.This thesis details the development of virtual reality training platform for EMG controlled prosthesis users that overcomes the drawbacks of the EMG training,making the process entertaining,less tiring due to the use of a virtual hand,and adaptable and progressive for a faster adaptation.This platform consists in different parts.First an application,made with the software Unity3 D and coded in C# integrates a virtual hand model as well as a game environment.The hand model has been coded so that the mechanical features putting the model into motion are receptive to external inputs in the form of discrete command through serial port communication,overcoming the known issue of virtual grasping.For the purpose of the study,this input have been set as the EMG armband enabling the user to control the virtual hand with the residual limb muscles signals.The platform also includes feedbacks,in the visual form,by indicators coded in the application as well as the implementation of virtual reality,and by the integration of vibrotactile feedback by the use of a microcontroller and vibrators.The study also explored different possible inputs such as Leap motion or data glove and it was discovered that these control could also be used for post stroke rehabilitation.Concerning amputee rehabilitation,the application fulfill the listed requirements and the platform could be used and continued to be developed by the creation of more applications.