Advanced search
Start date

Innovations in embedded navigation and computer vision for mobile robots


The use of mobile autonomous robots in social settings depends on the robot's ability to perform tasks required people to interact there. This ability begins with the autonomous navigation without shocks, predictable in dynamic environment and with respect to social movement regulations. The mobile robots are important in many service applications [2] [6] [7] [21], some markets now account for billionaires, among which we can mention: * Moving loads in industrial and logistic environments;* Surveillance and inspection environments; * Unmanned Aircraft; * Mobile services in hospitals; * Telepresence robots and audio-visual monitoring of convalescent; * Monitoring of the elderly in their homes; * Support in dangerous or inaccessible environments; * Household devices driven routine tasks. The company's mission Cybox Technology is to integrate the new technologies of robotics and embedded systems in the areas of security and survey. In order to meet a demand in the security industry, monitoring and inspection we developed the autonomous vehicle AutoDrive, mobile monitoring equipment. To learn more about the product please visit: security market grew 13% and electronic security 20% in 2014 (source: ABESE, 2014). This market already incorporates technology solutions and seeks to reduce operating costs. Demand from this market was proven by sales calls and expressions of interest. Tests were performed at Braskem Paulinia, in the parking Shopping Indaiatuba, in logistics condominium in Indaiatuba, in the Alpes and Itaici and Industrial plants in Guarulhos region. Customers have suggested a number of changes, some of which were made with company resources, such as increasing the number of cameras, the geo-system duplication, GPS, to fit the standards of safety equipment, the preview availability of Web cameras, 3G and WiFi, increased weight and size of the equipment, increased autonomy and contracting insurance for the equipment. Other suggestions, which received special attention from the engineering are state of the art navigation of autonomous vehicles and computer vision for embedded systems. Such developments led to the submission of this proposal to FAPESP. The main technological innovations suggested by customers and listed below belong to the areas of autonomous vehicles and smart sensors:* Reduction vehicle navigational to centimeters* Reduction indoor vehicle navigational error * Detection of people and cars* Detection guides and lines* Automatic loading equipment this proposal PIPE - Phase 1 research and development is divided into three modules: Navigation, Vision and Charging Station. Phase 1 will also be investigated and detailed the methods and algorithms to be embedded in the PIPE Phase 2 project. In the Navigation module seeks to integrate the method of geo-referenced differential navigation, DGPS for error correction cinematic real-time RTK, [12] and use the Particulate Filter, FP, for sensor fusion [51], whose entries the corrected signals from RTK DGPS and inertial system information provided by the IMU, IMU, [8] with 10 degrees of freedom (DoF 10).The objective is to view module board a person detector system, cars, guides and tracks, we will investigate the descriptor method of characteristics based on histogram of oriented gradients, HOG, [2] [3] and the algorithm (machine) classifier Vector , SVM, [3] using the OpenCV library in graphics processor, GPU, Mali 600 [5] with the Ubuntu Linux operating system. The innovations proposals are critical to sales success. (AU)

Articles published in Agência FAPESP Newsletter about the research grant:
Articles published in other media outlets (0 total):
More itemsLess items