IMUs Help Deliver the Daily Bread

July 1, 2020

Tony Zarola

The term ‘autonomous vehicle’ often conjures the image of driverless cars, but a more tangible and nearer-term application of vehicle autonomy is with the “farm-to-table” movement that promotes serving fresh, locally-grown food at restaurants and school cafeterias.

The new “normal” necessitated by the COVID-19 pandemic created a surge in home delivery of groceries and all kinds of products. This increased demand, coupled with restrictions such as social distancing, has boosted usage of autonomous machines, which have quickly become integral to grocery stores, warehouses and hospitals. Autonomous machines touch practically every aspect of the farm-to-table ecosystem, from agricultural equipment used to harvest produce in crop fields to semi-autonomous commercial trucks ferrying goods to warehouses, and finally, last-mile delivery vehicles including vans, drones and automated ground vehicles (AGVs).

Though the degree of autonomy may vary from partial (trucking) to full (AGVs), at the core of all these systems, the inertial measurement unit (IMU) is proving to be a critical component in their synchronous and uninterrupted operation. For operations such as tillage, planting, chemical and fertilizer application, and harvesting, IMUs ensure machine accuracy and orientation sensing through difficult terrain. This allows vehicles to continue working through complex motion profiles, including rotations in multiple directions and experiencing shock, to keep equipment on-course and on-task. This maximizes crop productivity and yield, while minimizing time, labor and material inputs.

All-weather performance

Consider a loaf of bread and its journey from the field to your dining table. Once the grain arrives at the manufacturing facility, it is cleaned, sieved and milled, increasingly by robots and cobots (collaborative robots that work alongside humans) that are familiar applications of autonomy. Beyond machines that perform tasks on a conveyor belt, warehouses currently use lift trucks and belt systems with varying degrees of automation to move or organize materials. AGVs equipped with IMUs in combination with perception or magnetic sensors that form a precision navigation also are used to efficiently move goods within a geofenced area, allowing orders to be filled faster. The finished flour then is baked into bread ready for the dining table.

The next automation touchpoint is the transportation of the finished product (bread) to the grocery store shelf and on to its final destination: homes. This is accomplished through a combination of semi-autonomous commercial trucks, grocery store robots and/or last-mile delivery vehicles including vans, drones and AGVs that drop your fresh groceries and that loaf of bread at your doorstep.

As companies adopt and implement these technologies faster than ever to ensure their employees’ safety and wellness, there has been a corresponding push to drive rapid advancements in the performance and accuracy of IMUs.

Currently, automation in vehicles and machines is enabled by an interdependent, intricate system of sensing and perception modalities, including GPS, cameras, radar, lidar and IMUs. All of these need to work in near-perfect harmony to enable fully functional, fully autonomous machines.

However, other than IMUs, each of these modalities comes with its own drawbacks and limitations that impact performance and uptime. The IMU, which features a multi-axis combination of calibrated precision gyroscopes and accelerometers, remains unaffected by environmental factors such as fog, snow and rain, rapid temperature changes, loss of signal when going under a bridge or through a tunnel and vibration feedback due to an uneven or bumpy driving surface.

IMUs are a plug-and-play technology that reliably senses and processes multiple degrees of freedom, providing a consistent signal when other modalities are compromised, leading to information lags and gaps in the fusion engine. It is during these critical ‘flying blind’ moments that the machine receives and relies on the speedier data supplied by the IMU, helping the vehicle to continue on its intended path, or bring it to a safe halt if the system determines that safe travel is not possible.

Predicting bias error

With enhanced performance and sustained immunity to environmental and system interferers, the IMU bias-error growth can be better predicted – this not only will help overcome the significant and potentially dangerous lane drift/variance during outages of other sensors, but also enables safe operation of the vehicle as speed envelopes increase beyond the current limitations. Such performance is engineered into ADI’s evolving IMU portfolio, including the recently introduced compact ADIS1650x family that features robust improvements in dynamic alignment (or cross axis sensitivity), vibration rejection and bias stability. These features enable greater positional accuracy of autonomous navigation systems.

This higher performance – i.e. lower latency, improved accuracy and greater speed of operation – will lead to a shift in the role of IMU in autonomous vehicles. From a dependable back-up player, we anticipate the IMU to assume a more central role in the navigation system, where it will arbitrate input from the various sensing modalities and determine which signal is the most accurate and reliable.

IMUs feature full factory calibration, embedded compensation, sensor processing and a simple programmable interface. High performance IMUs have the potential to make autonomous vehicles safer, more reliable and faster for all users. As the world continues to adapt increasingly to automation, accuracy and trust will continue to be important factors in driving adoption. From manufacturing facilities to cars on the highway and farm equipment in the field, the reliability of IMUs will help push the envelope of what’s possible with automation.

---