Positioning to Centimeter Level Improves Ag Productivity

Precision navigation breakthroughs and approaches that emphasize backward compatibility help farms reap more value from autonomous operations.

Inertial navigation systems use GNSS and RTK data to guide autonomous and semi-autonomous vehicles. (ACEINNA)

Precision navigation, which is essential to autonomous machines, is knowing a moving vehicle’s absolute and relative position in 3D space accurately, reliably and repeatably. The cost for precision navigation with position information down to the centimeter level has become a far less significant barrier to broad deployment. This price drop has come with enhanced, not diluted, performance.

Comparison of IMU/INS application, performance, size and cost. (ACEINNA)

One reason for the cost decrease has to do with the sheer volume of Global Navigation Satellite System (GNSS) receivers on the market. GNSS encompasses the satellite-based GPS used in North America as well as other global positioning systems such as GPS, Galileo, Glonass, BeiDou, NAVIC (former IRNSS) and QZSS.

Because there are no lane markings, signs or HD map access in a wheat field to guide a tractor, farmers often need to rely upon GNSS satellite-provided guidance. However, traditional GNSS only provides meter-level accuracy. This is where the use of Real Time Kinematics (RTK) to enhance the accuracy of GNSS to the centimeter level comes into play. Benefits from applying RTK to GNSS navigation include minimizing seed waste through precise spacing between rows and even between seeds in the same row, more controlled fertilizer application to prevent both over- and under-fertilization, as well as minimizing equipment crop damage during growing season operations and harvesting.

High-performance MEMS IMUs

RTK improves GNSS accuracy in a two-step process. First, an RTK technique is used to measure the GNSS carrier signal and so obtain precise time information. Next, to further improve accuracy, RTK corrects the measurement using data from a network of fixed base stations.

Inertial navigation systems (INS) use GNSS and RTK data to guide autonomous and semi-autonomous vehicles. The main subsystem of an INS is an inertial measurement unit (IMU), which measures acceleration and rotational rate. It calculates velocity, relative position and attitude. Accelerometers in the IMU measure linear acceleration in all three dimensions, while IMU gyroscopes measure angular (rotational) rate about the x, y and z axes. In this way, the changes in an autonomous tractor’s velocity, position and attitude, for instance, can be determined.

Higher-performance micro electro-mechanical systems (MEMS) sensors are now available at a cost and size smaller than the gimbal and fiber-optic gyroscopes once used for IMUs. The arrival of MEMS on the scene is making precision navigation more practical for agriculture.

High-performance MEMS IMUs have steadily improved the cost-to-performance ratio. Once they weighed in at a hefty 500 grams (17.6 oz) and measured 100 x 100 x 100 mm (3.9 x 3.9 x 3.9 in.) at costs exceeding $500 each. Today, high-performance IMUs can weigh and measure as little as a quarter or a euro coin and can cost less than $30.

INS meets off-highway challenges

Machinery in agricultural environments face navigation challenges that on-road vehicles do not. Besides the absence of marked lanes, HD maps and directional signs, there are also equipment characteristics to consider. For example, solutions for safe navigation and movement by farm equipment must take vibration into account.

In the presence of vibration, an IMU’s signal chain must be on its toes. It must protect the integrity of the signal needed to steer a vehicle or safely operate an implement such as an automated seeder or hay baler. Rejecting and eliminating vibration-related acceleration and vibration-induced error is vital.

The most effective IMUs are calibrated on precision rate tables and in temperature chambers. This approach preserves accuracy and consistency over input range and temperature.

Meeting ISO 13849 with backward compatibility

James Fennelly, product manager – Inertial Measurement Systems, ACEINNA Inc. (ACEINNA)
IMUs can successfully measure movement in three different axes. (ACEINNA)
The MTLT335D Dynamic Tilt Sensor Module provides pitch, roll, 3D acceleration and 3D rate sensing. (ACEINNA)

Safety is a common factor that is critical for all agricultural machinery. The need for safety runs through autonomous applications as diverse and sophisticated as weed killing to increase crop yield, avoiding over- or under-fertilization or boosting irrigation efficiency. As automation grows, machine safety must keep pace. This is why systems and subsystems that meet the ISO 13849 safety standard have begun to arrive on the market.

One example of a subsystem used in autonomous farm machinery, and which meets ISO 13849 Category 2, Performance level d (PLd), is the MTLT335D Dynamic Tilt Sensor Module from ACEINNA. From the ground up, MTLT335D development followed the design and safety requirements detailed in ISO 13849.

Keeping the same form factor, electrical interface and connector as the previous generation (MTLT305D) preserved backward compatibility even as key performance parameters improved. For example, without changing the form factor, two key parameters related to gyro performance, Angle Random Walk (ARW) and Bias Instability (BI), were improved by a factor of three.

The lower the ARW number, the less error there is during the integration of the rate signals used to calculate attitude and position. BI is a measure of how much the gyro bias drifts with time. Improved BI enables quicker and better estimation of true gyro bias and results in more accurate attitude and position estimations.

The MTLT335D module’s system wiring harness and SAE J1939 CAN interface messages are the same as the previous-generation MTLT305D. Keeping these the same allows the system to use the same software to decode measurement messages, largely preserving system software compatibility. By making migration to ISO13849 straightforward, OEMs and farm equipment manufacturers can improve their solutions without requiring significant redesign.

Where is farming going next?

“Anywhere it wants” is the glib answer to that question. With just a cursory look at today’s agriculture, that answer doesn’t seem far from the truth. Look at what agriculture is achieving as advanced technologies become more affordable, adaptable and agile. Terrain once too uneven to be considered arable is being conquered with sensors that calculate and correct a tractor’s roll, pitch and yaw.

Autonomy is freeing farmers from the tedium of tractor guidance and is making their farms much more productive and efficient. Water, fertilizer and pesticide resources can now be applied with precision that is approaching individual plant level, shrinking waste and expanding productivity.

James Fennelly, product manager – Inertial Measurement Systems, ACEINNA Inc., wrote this article for SAE Media. Fennelly received his BS EET from the University of Massachusetts and has been working for the past 15 years with MEMS inertial sensors including component-level acceleration sensors and system-level products.