in Canopus Blog
MEMS INS GPS : The Ideal Navigation Solution For Robotics Applications
Use of robots for domestic as well as commercial purposes is growing rapidly. As the cost of robots goes down and their versatility increases, a lot more tasks, and chores would be performed by robots. Precise and accurate navigational assistance is absolutely essential for robots to function properly and carry out their ordered tasks. This calls for a navigation solution which is accurate, give fast data and is cost effective. Using GPS for navigational assistance is not possible always as robots may operate indoors, GPS signals could either be not available or inhibited. Also, in order to control movements of the robot and make it travel along a designated path, the motors need data at a higher rate than is provided by a GPS.
Along with the above-mentioned limitations, in the indoor spaces, GPS provides fairly accurate two-dimensional navigation assistance which is not adequate for certain applications like robots working in a multistoried building. On the other hand, using sophisticated RLG or FOG based inertial navigation systems is not commercially viable due to the high costs of the sensors.
Numerous experiments and research conducted over the years hint that a combination of solid-state MEMS inertial navigation system with GPS or Odometer / Encoder aiding effectively solves all of the above-mentioned problems. A MEMS inertial navigation system with a magnetometer, odometer input, and integrated GPS is affordable and provides a precise and accurate output. Since it is a hybrid system, it can function properly in the circumstances where the GPS signal is the week.
In summary, if compared with all other navigation options, MEMS inertial navigation systems with integrated GPS have the following advantages:
- Higher data output rates: MEMS inertial navigation systems have higher output rates when compared to pure odometer or GPS the only solution. Sensor sampling time should be at least twice (in practice 4 to 5 times) response time of actuators for effective control which MEMS INS can deliver.
- Lower power consumption: MEMS sensors since they are essentially micro-electronic devices consume a lot less current than laser or optics based sensors. Typically MEMS IMU/INSs consume 2-5W power whereas FOG INSs consume 10-30W.
- Orientation monitoring and control: Standalone GPS systems cannot monitor and provide information about the orientation of the platform in static or semi-static conditions. MEMS INS comprising magnetometer and gyroscopes provides orientation information about platform orientation and can suitably prompt motors to control it to prevent it from becoming unstable.
- Cost effective: MEMS IMUs/INSs are low cost ($1000 – $2000) which is justifiable for industrial or consumer robotic applications. Compared to FOG or RLG systems the cost of MEMS inertial navigation systems (with integrated GPS) is 40-50 times less.