What Robots Can Learn From Your Tesla

Tesla has disrupted the traditional car manufacturing industry in many ways, and its use of data is no exception. Tesla's cars are equipped with a vast array of sensors and data collection tools. This data is then transmitted to Tesla's servers, where it is analyzed to identify potential issues, optimize performance and improve the user experience.

One of the significant ways that Tesla uses data is through over-the-air updates. Unlike traditional car manufacturers, that require customers to bring their vehicles into the shop for software updates or recalls, Tesla pushes software directly to the cars via a wireless connection. These updates can include improvements to performance, new features and bug fixes, all of which can be deployed quickly and efficiently. The impact is that cars improve over time, helping them maintain their value.

Sensor data from many millions of miles driven by Tesla owners is also used to develop the controversial Full Self-Driving option. In addition, Tesla uses data to provide their customers with a range of services that traditional car manufacturers are struggling to catch up to. Tesla's remote diagnostics feature allows the company to identify issues and proactively schedule service appointments. This can save customers time and money, as well as reduce the risk of more significant problems developing over time.

This use of data represents a fundamental shift in the way that car manufacturers operate. By collecting and analyzing data in real time and using over-the-air updates to improve the performance of their cars, Tesla provides their customers with a better experience. As the automotive industry remains highly competitive, other automakers are chasing Tesla’s tail lights.

Some of these lessons can be applied more broadly to robotics. Autonomous robots are revolutionizing industries and changing the way work is done in areas as diverse as food, agriculture, supply chain and healthcare. Most of the attention on robots is around how they can augment—and in some cases replace—human labor. However, this underestimates the impact that robots may have. Beyond just doing the same job as people, robots are capable of both using and generating data at a level of detail that humans can’t even come close to, taking digital transformation from hype to reality.

Autonomous robots are equipped with a variety of sensors, not unlike those in modern cars, including different types of cameras (RGB, infrared, stereoscopic), 3D-sensing LiDARs, sonar and other data collection tools. Mobile robots can be used to carry around additional sensor payloads, such as to detect gas leaks or to listen for anomalous machine operation. These sensors collect information about the environment in which the robot operates, as well as information about the robot itself. Sensors capture the robot's location through a process called SLAM, as well as information about its battery charge and the objects around it.

Data is then processed in real time by the robot's onboard computer systems, which use machine learning algorithms to analyze the data and make decisions about how the robot should behave. This may include defining a path to reach a goal while avoiding objects in its environment, such as a group of people or a misplaced pallet.

An even bigger impact is the ability of robots to produce valuable data about the environment in which they operate. As a robot carries out its tasks, it generates a constant stream of data. Aggregate data collected over time, across a whole fleet of robots and throughout multiple sites, can be used to optimize every robot's performance, improve their accuracy and enhance their ability to make decisions. Data can also be leveraged to improve overall operational efficiency beyond the robot itself, by providing a self-updating map and correlated data.

Consider cleaning the floor in a large public area like an airport terminal, which can span millions of square feet. Traditionally, human workers will follow a rough schedule and may capture some high-level information about their work, often filling out a timesheet. By comparison, an autonomous floor scrubber can capture the last time that each square foot was cleaned, how much detergent was used and any areas that were missed. This results in verifiable cleanliness.

The data generated by autonomous robots can also be used to improve safety and reduce the risk of accidents. A robot that is used for inspection in a hazardous environment can use data about the temperature, pressure and other factors to identify potential safety risks. It can then adjust its behavior or alert human operators to take action to mitigate those risks.

Robots are also voracious data consumers. Smart software in the cloud can direct robots used for logistics and warehousing—using data about the layout of the warehouse, the location of items, the amount of inventory, and even the weight and size of items—to optimize the robots’ path for picking and moving items. This results in faster and more efficient processes.

Combining the robots’ onboard data with external data from fixed sensors, such as ceiling-mounted cameras, smart doors and occupancy sensors, can help eliminate blind spots and also influence the actual robot behaviors. In the airport terminal example, the cleaning robots may respond dynamically to actual usage, cleaning low-traffic areas less frequently or increasing the amount of water used in areas that need extra cleaning.

Using advanced orchestration software, different robots can work together to carry out missions, even if the robots are made by different vendors. Orchestration can go from basic co-existence to clever coordination, improving efficient operations and ultimately collaborating on complex tasks that no robot can complete on its own.

Robots can bring the kind of data-centric experience introduced by Tesla to every physical industry. In an era of heightened awareness about efficiency, supply chain disruptions, increased geopolitical instability and the lowest unemployment rate in 50 years, companies must look beyond minor incremental improvements and embrace a more data-driven approach.

* Photo credit - Tesla.com
* This article was originally published on Forbes.com