All the news that's fit to assimilate[ Home | Blogs | Events | Robots | Humans | Projects | Podcasts | About | Account ]
Click on an image to enlarge it
|Target Environment||Locomotion Method|
|Sensors / Input Devices||Actuators / Output Devices|
NTSC color camera
micro-switch grip sensor
|2 hacked RC servos
3 standard RC servos
|Control Method||Power Source|
|CPU Type||Operating System|
|Time to build||Cost to build|
|URL for more information|
|This robots name is 'Mars Rover 1'. Is a major rebuild of a previous robot called 'Monitor'. This robot uses servos to tilt a camera and sonar scan platform to allow the operator to see obstacles and objects of interest. The tillable gripper has a pressure switch to indicate to the bot its grabbed an object.
The premise of the robot is to simulate the difficulties of remote control at planetary distances. A control station using a pic for user inputs inserts a 10 second delay between communications with the robot. The station uses a key pad and converted joy stick for control inputs. A sequence of actions are entered and then 'Up Linked' to 'Mars'.
The resulting simulation showed it to be very difficult to run a rover. In this instance the robot with its deferential tank tread had very non linear responses to the carpet, simulating the surface of mars. A series of control commands would not generate a predicable movement due to the skiddering motion of the robot, worst when turning. The result, without being able to have a 'gods eye' view of the bot but only though its camera and sensors via TV and the LCD on the command console the operator could become very confused and disoriented as to what the robot did and where it was!
Problems were also encountered with the weak digital RF transmitters. The robot had to be within about 30 feet, just on the other side of a wall, to keep the operator from seeing the robot.