Venom A Climbing Robot For Metal Walls

The Robot presented from University of Catania to the Clawar 2004 Student Competition is named ”Venom Robot”. This particular type of Climbing Robot has been realized after approximately 6 months of hard job that was divided between the development of the mechanics of the robot and its adhesion methodology, the choice of suitable motors to climb the wall, the choice of a suitable control unit, the realization of same sensors, and finally the development of suitable software for robot navigation (autonomous and tele-operated mode). The target of the Venom robot is to climb a vertical metal wall, from bottom to top, without any human interaction and to navigate around obstacles of unknown size and location. The robot stay attached to the wall by means of a Magnets Plane located on the bottom of the robot.

The first step for the Venom Robot realization was the development of the body of the robot. The design of the robot body has been done in SolidWorks CAD environment. The body is realized with aluminium and its weigth is of 2.1 Kg (without battery). There are two gearboxes between the DC motors and the wheels; both have 1:2 ratio. The Venom robot has two DC motors that are the Escap 26N 58 113 R22 65.5, where 65.5 is the ratio of the integrated gearbox. The H-Bridges allow us to control the motors speed and direction by means of a PWM signal and a direction bit; the H-Bridges used are the TPIC0107B produced by the Texas Instrument. The Microcontroller Card used in Venom Robot is the Infineon C167CS that together with the Development Board phyCORE-167 allow the control of the all robot’s components. The architecture of the C167CS combines advantages of both RISC and CISC processors and of an advanced peripheral subsystems in a very well-balanced way. In addition the on-chip memory allows the design of compact systems with maximum performance. We have realized seven optical sensors to allow the robot to localize the obstacles on the wall. These sensors rely upon the Sharp IS471F photo-transistor and a Infrared light emitting diode (λp= 940nm). There are two LEDs (red and green) that are used in presence or absence of obstacles. We have realized an inclinometer to allow the robot to know its orientation on the wall. This sensor relies upon the Analog Devices ADXL202JE 2g Accelerometer. We have used a Logitech Optical Mouse that, with some modifications, became a very good relative position sensor. It has two outputs channels X and Y that supply two TTL quadrature pulses corresponding to the robot relative position.