No abstract

The research project "SafeCon" tackles the problem of the safety in military transport missions. In such missions people can be killed by attacks or natural dangers, for example avalanches. To reduce the number of involved people in transport missions, SafeCon tries to reduce the number of requied personnel by automating each transporter.The aim of SafeCon is the realisation of a semi-autonomous truck convoy. The destruction of one vehicle within the convoy causes financial losses and spares human lives. Further, the vehicles must not be armoured. This reduces both load capacity and costs. For an autonomous behaviour, each truck must be able to follow the vehicle ahead. Because of that, each vehicle must be able to recognise the leading vehicle. This is realised with a computer visiondetection system. To enable an autonomous behaviour in all weather conditions a thermal infra-red camera system is implemented to detect objects based on their heat signature. Different to other approaches, this paper discusses the implementation of an AdaBoost based detection for thermal infra-red image material. The evaluation of our approach compares a thermal infra-red and a visible-light detection. Ourexperiments prove that the thermal infra-red detection is comparable with visible-light detection.

No abstract

To autonomously navigate through unstructured environment mobile robots rely mainly on vision systems. Commonly used are stereo vision, laser and infrared sensors. This paper deals with an evaluation of the Kinect depth sensor and compares it with stereo vision. Also the depth resolution, the accuracy and the repeatability have been determined throughout several experiments based on the SDK of Code Laboratories (CL) for Windows 7. Our experimental results show that the Kinect has a maximum depth resolution of 1.3[mm/color value] and a maximum accuracy of 0.02%. Also, depth density has been assessed and compared to stereo vision images, concluding several benefits the Kinect brings to the field of mobile robotics. Finally, suggestions are being made for several applications for use in mobile robotics

No abstract