Sensory system in autonomous robots


A tele-supervised autonomous robot needs to have accurate information on its position, orientation, and velocity and the location of nearby obstacles (sensory system). The robot needs multiple sensors to gather all the information required and a computer to read the sensor data and make decisions based on it. Information on the robot’s environment can be acquired using laser distance sensors, ultrasonic distance sensors, infrared distance sensors or cameras utilizing machine vision. Other methods are also available, but they are not discussed here further. The most popular detection methods in robotic projects have been using a laser range finder and/or a computer vision system. Some studies have also used infrared or ultrasound sensors as alternative obstacle detection sensors. An intelligent robot has to have sufficient knowledge concerning the state of the outside world. For this it needs many sensors (aka sensory system) to gain information regarding its environment and status. To effectively use all of its sensors, a method is needed for integrating the information from different sources into a more usable form for the operating system.

Sensor Fusion

In sensor fusion, the raw sensor data from the sensory system is transformed into a more abstract numeric or symbolic representation. In robotic applications, it is important to detect and study the kinematic of objects. The movement or the displacement of an object is directly related to the position i.e., the object coordinates about a reference system. The displacement of an object in space from an initial position to a final position can be decomposed into linear displacement or angular displacement. The linear displacement refers to the distance between the initial and final position, and the direction of travel in the direction of travel, i.e., the straight line on which it is moving. The angular displacement refers to the angle between the initial and final position of the object and the direction of travel on the plane of movement. For that reason, the following sensors provide the minimum selection of sensors which can be used to provide the economic thus effective way of relative position detection system.

Proximity Sensors

Proximity Sensors are used to detect an object regardless of the distance from the sensor. These sensors are divided into two major categories depending on the mode of operation, detecting by contact or by measurement.

  • The proximity sensors based on the contact of the sensor to the object, detecting whether an object is close enough to the robot then activates the appropriate robot circuit. The objects are not in contact with the sensor are ignored.
  • The distance measuring sensors calculate the distance between the sensor and the object lying within the range

The contact is the most common form of object detection. They are usually present in a completed sensory system. The contact sensors help a robotic system to perceive the contact or not with an object. Implemented using switches that change state when there is mechanical contact. The form is simple and economical, but with multiple uses. The switches are widely available in a wide variety, and easily connected to robotic systems control systems. This platform uses 5 contact sensors to control the connection between the floor and edge of each leg. So, it gives the opportunity to move on uneven level, reduce energy consumption and prevent a problem such as a multitude extensive pressure specifically feet from incorrect weight distribution. Below is a presentation of the legs of robotic platform with sensors, and some cases of motion.

Inertial Navigation Systems

The Inertial Navigation Systems (INS) are integrated sensors devices which record the acceleration and the rotation rate of a system of three axes, assisted of three accelerometers and three gyroscopes. Comparing GPS and INS notice the following. The INS do not require and do not depend on an external electromagnetic energy signal. So their use is feasible in areas outside of GPS range, eg inside buildings. However, the INS have problems due to the integration of the signals, a process that accumulates measurement errors. The amount of data generated by the sensors requires substantial computational resources to process them. The INS systems used in passenger aircraft, where the accumulation of error in measurements reaches 1800m per hour of operation. The maximum achievable accuracy approaching 0.1% of dianystheisas distance in terrestrial applications. The construction of a platform in which to maintain a constant orientation of the accelerometers system used gyro balancer device is very complex and very costly, which is prohibitive in most cases for applications in robotics. Nevertheless the progress of laser devices and optical fibers greatly reduced the cost and use of INS systems are now more affordable for applications in robotics.

The Intertial Measurement Unit (IMU) plays an important role of inertial navigation systems. The IMU inertial measurement unit is an electronic device that takes measurements of speed, the orientation of a vehicle and the gravitational forces, using a combination of accelerometers, gyroscopes and magnetometers sometimes.In that capacity, the data collected by the IMU sensors allow a computer to monitor the position of a vehicle, etc., using the calculation of their position.The IMU devices used in vehicles, boats, aircraft, unmanned aerial vehicles, spacecraft, satellites and equipment landing. An IMU device allows the GPS to operation when the signal is not available, such as when there is electrical interference. For example, supposingly an IMU device has been installed on a plane, when it finds that the vessel travels to the east for an hour with an average speed of 500 miles per hour, then the guidance computer will assume that the plane must be 500 miles east of the original place. In combination with an electronic map system, the steering system can be used to indicate to the pilot where is the plane. It is similar to a GPS system but without the need to communicate with external
elements, such as satellites.

Disclaimer: The present content may not be used for training artificial intelligence or machine learning algorithms. All other uses, including search, entertainment, and commercial use, are permitted.