The developments that took industrial robotics from automation to collaboration.

Industry 4.0 Deep Dive. Part 4: Evolving Industrial Robots

<p>All equipment and machinery across the EU is standardised through European Commission directives, including wiring. These directives are referred to as EN (European Norm) codes and are usually formulated in compliance with standards set by the International Organization for Standardization. At Rowse, we are able to supply a wide variety of control cables for wiring your control panels and systems, but many of our customers are uncertain about the rules governing colour codes.</p><h1>Wiring Colour Codes</h1><p>The colour codes used in the past were originally determined by the British standard regulations BS 7671, but these standards were set down before complex automated systems were even thought of &ndash; and they largely related to domestic circuits and static installations. The EN codes supersede the old British Standards (for the time being, at least) and the international standard EN 60204-1 specifically governs Control Panel Wire Colour Codes.</p><p>EN 60204-1 (Safety of Machinery &ndash; Electrical Requirements of Machines) sets out both the required safety standards and the type and colour of cable that should be used for wiring an electrical enclosure. In addition, the International Electrical Commission has also issued a revised standard IEC 60445:2017, covering safety principles for HMI terminals, including the marking and identification of equipment terminals, conductors and conductor terminations.</p><h2>CONTROL PANEL WIRING COLOUR CODES PER EN 60204-1</h2><p>The choice of cable colour initially depends on what type of circuit it is, and whether the voltage is AC or DC.</p><p><strong>AC and DC Power Circuits</strong></p><table><tbody><tr><td>AC Phase and DC + -</td><td>Black</td></tr><tr><td>AC Neutral</td><td>Light Blue</td></tr><tr><td>Protective Conductor (Earth)</td><td>Green and Yellow</td></tr></tbody></table><p><strong>AC and DC Control Circuits</strong></p><table><tbody><tr><td>AC Live or Neutral via transformer</td><td>Red</td></tr><tr><td>AC Neutral (mains)</td><td>Light Blue</td></tr><tr><td>DC+ -</td><td>Blue</td></tr><tr><td>Earth</td><td>Green and Yellow</td></tr><tr><td>Interlocks control circuits supplied from an external power source or remote panel</td><td>Orange</td></tr></tbody></table><p><strong>Panel Services Colour Coding (multicore)</strong></p><table><tbody><tr><td>AC Live or Neutral cores</td><td>Orange</td></tr><tr><td>Earth cores</td><td>Green and Yellow</td></tr><tr><td>Sheath</td><td>Orange</td></tr></tbody></table><p><strong>Panel Services Colour Coding (single core)</strong></p><table><thead><tr><th>Type of Circuit</th><th>Colour</th></tr></thead><tbody><tr><td>AC Live</td><td>Orange</td></tr><tr><td>AC Neutral</td><td>Light Blue</td></tr><tr><td>Earth</td><td>Green and Yellow</td></tr></tbody></table><p>The wire used for the protective conductor (earth) must be GREEN-AND-YELLOW. This colour combination is reserved specifically for the protective earth and must be maintained throughout the wire&#39;s entire length. If the wire is insulated, the GREEN-AND-YELLOW combination must be a 70/30 distribution over the surface area on any 15mm length (this is usually produced as GREEN with a YELLOW stripe). If the wiring is installed in an inaccessible position or is clearly identifiable as the protective earth, the GREEN-AND-YELLOW colour coding can be confined to any location where the wiring is accessible or to the ends of the conductor.</p><p>The specified colour for the neutral conductor wiring is LIGHT BLUE. This colour should be restricted to the neutral wire and not used for any other associated cables, so that there is no possibility of confusion between neutral and some other function. This colour should permeate the entire length of the neutral conductor, or if the wiring is in a concealed location or somewhere with limited access, the cable can be identified with a 15-100mm LIGHT BLUE stripe.</p><p>You should use neither GREEN nor YELLOW Control Panel Wiring separately where there is any possibility of it getting confused with the combined GREEN-AND-YELLOW wire used for the protective earth.</p><h2>Tri-Rated Cable</h2><p>The very popular Tri-Rated Cable (rated under BS 6231) is a kind of high temperature, fire retardant cable specifically designed for control panels used in power switchgear. It is also rated for use by the standards organisations of the US and Canada, hence the &#39;Tri-Rated&#39; title, and can also be called control panel wire, switchgear cable or tri-rated single (TRS) wire. Tri-Rated Cable can function at temperatures of up to 105&ordm;C, but for safety&#39;s sake BS 6231 stipulates 90&ordm;C as the maximum operating temperature. Conversely, the British Standard voltage rating for the cable is higher than the voltage permitted in the United States equivalent.</p><p>The advantage of the triple rating is that it can be used in a much greater range of markets, reducing design duplication and allowing for a much greater degree of flexibility in manufacturing processes. The panel wire is flexible, with a single core and one layer of heat-resistant PVC insulation. It has varied applications in many industry sectors, being used for wiring electrical cabinets, instrumentation panels or relays. It can operate inside small electrical devices, and also functions as an internal connector in things like rectifier equipment, motor starters and controllers.</p>

Standardisation is vital for the safe installation of electrical components, and even more so when machines are becoming so sophisticated.

Control Panel Wiring Colour Codes

<p>This digital representation technology enhances the already strong links that tie the digital world to reality, and provides an even closer representation of how things work than 3D simulations. Digital-twinning is a holistic approach to the many and varied processes that go into modern industry, and demands a quantum shift in production culture.</p><h2>What Are Digital Twins?</h2><p>Digital twins are a complex conglomerate of technical data, combined to make a virtual technological representation of a physical process or product. They simulate a product&#39;s entire life cycle, predicting its behaviour, and optimising its design and production system. They incorporate AI and machine learning capabilities, data analytics and multi-physics simulations. They demonstrate the potential impact of environmental or usage conditions, design changes and many performance variables. Digital twins are a cost-effective means of understanding the performance characteristics of a real-world physical counterpart, before any investment is made in actual prototypes. This cuts down drastically on development time, and delivers a final process or product of optimal quality.</p><p>Obviously, it&#39;s necessary to use physical objects as the paradigm for digital twinning, in order to ensure that the technical modelling is accurate. This takes place over a product&#39;s entire lifetime or production cycle, using data gathered from sensors installed on such objects. This data will reflect real-time performance of the modelled object or process, including its operating conditions and any alterations in performance that occur over time. The digital twin continuously updates the model during its life cycle, constantly evolving to create a virtual feedback loop.</p><h2>Digital Twins Use Cases</h2><p>Digital twins generally fall into three categories of asset modelling. These are combined and integrated into a digital thread which permeates the entire fabric of the asset&#39;s life cycle.</p><p>Product digital twins consist of a virtual link to a model product that demonstrates how it performs in various operating conditions, obviating the necessity for multiple physical prototypes. Product digital twins also improve the manufactured product&#39;s final quality, and provide a much faster and more effective response to feedback.</p><p>Production digital twins simulate production processes, analysing and validating what is happening and why, so as to create a more streamlined production methodology that can be optimised for different conditions. Sensory data can be used for predictive maintenance, making manufacturing operations faster, more reliable, and ultimately more efficient.</p><p>Performance digital twins capture huge amounts of operational data in a product&#39;s performance, analysing it with big data analytics and acting on the information to enable more informed decision-making.</p><h2>Digital Twins Vs Simulations</h2><p>Advanced 3D simulations are definitely an integral part of the new technology, but digital twinning goes far beyond that. Digital twins are a comprehensive concept incorporating multiple other technologies, including AI, the IoT, big data, edge computing and much more. All these combine together to construct a digital equivalent of a physical asset or entity, as an accessible 3D visual simulation. This allows human operators to monitor, analyse and respond to real time situations.</p><p>Simulations in the past helped designers to understand what could possibly happen to a product&#39;s performance in the real world, without being able to affect it. Digital twins examine the entire product life cycle, helping designers to understand not only what might happen to it in performance, but what is happening to it live.</p><h2>Digital Twins In The Rail Industry</h2><p>Digital twins incorporate the actual running of an existing rail network, enabling engineers to examine every step of the process as well as the equipment itself. With these networks still needing a lot of overhaul, the process must be cumulative, with a vast amount of different data needing to be coherently amalgamated.</p><p>Some operators want to use digital twins to provide full visibility of information throughout the design cycle, so that end users may have early design inputs to infrastructure, assets and projects. Digital twinning combined with augmented reality software would supply real-time asset simulations with live data. This thread could also incorporate predictive models of various system interventions, taking social outcomes into account to improve decision making. Digital twins allow real time changes to be recorded across networks, demonstrating the impact of human and environmental changes. They could be used to map out potential escape routes and human traffic behaviours in the event of a disaster.</p><h2>Digital Twins In The Military</h2><p>Military platforms in all domains function on multiple advanced technologies, such as communications and IT networks, command and control systems, surveillance, weapons systems, and much more. Design engineers must also integrate multiple different systems built by different manufacturers for many different purposes, with equipment becoming ever more complex and technologically advanced. Performance testing in varying conditions can be carried out using digital twins. This will help predict the platform&#39;s interoperability and integration issues, as well as scheduling maintenance and any necessary retrofitting.</p><p>Leaders want expensive fleets to last longer, but also to stay up-to-date with the changing demands of a service life that may extend for 50 years. Upgrades and retro-fitting are inevitable, and digital twins offer the opportunity to test and optimise designs before any expensive installation takes place. The functioning of complex systems and high grade materials can also be realistically simulated, to explore their performance range and the impact of changing roles. In extreme circumstances, digital twins could provide experts at base operations the means to help resolve problems down range.</p><p>The US military is already exploring digital twins with the IoT to see how all aspects of military ops can be connected, covering the whole cycle &ndash; from initiation, through maintenance, to eventual retirement. This holistic approach includes not only the fabric and systems of equipment, but troop movements, soldiers&#39; uniforms, and their health status. Constant data input from surveillance and other assets provides real-time insights into potential threats, including movements of enemy troops and vessels, so that more informed decisions can be made.</p><h2>Digital Twin Tech And The Future</h2><p>Many branches of industry are now looking into digital twins as the future of testing, control and managing complex systems, where all problems can be resolved before any substantial time or money is invested in physical production. Adopting digital twin technology involves much closer communication between designers, engineers, managers and end-users, but the increased investment in planning is far less costly in time and money than physically modifying an existing product.</p><p>This post was originally posted at <a href="https://www.rowse.co.uk/blog/post/an-introduction-to-digital-twins" target="_blank">https://www.rowse.co.uk/blog/post/an-introduction-to-digital-twins</a>.</p>

Digital Twins are the latest crystallisation of the technology advances loosely connected under the umbrella of Industry 4.0 and the IoT.

An Introduction To Digital Twins

<p>Machine building has become more complex with the introduction of more automated systems, and designers must devote more careful attention to the newer safety standards that apply. Mechanical safety standards had to be supplemented with new procedures for functional safety and performance reliability. In 2006, <a href="https://www.rowse.co.uk/blog/post/iso-13849-performance-levels" target="_blank" title="ISO 13849">ISO 13849</a> was revised to incorporate safety functions for machine control systems, reflecting the move towards more software-based architecture.</p><p>Designers must anticipate every possible eventuality for the system, whether it is operational, not operating or being serviced. Incorporating standards like ISO 13849 will ensure the maximum possible safety of pneumatic circuits and components, based on risk assessment and performance level (PL) parameters. Using factors such as MTTFd (mean time to dangerous failure), CCF (common cause failures) and DC (diagnostic coverage), the PL can be calculated for each individual safety function, and a required level of performance stipulated (PLr).</p><h1>5 Key Safety Functions in a Pneumatic System</h1><p>When designing a pneumatic system, there are five key safety functions that must be considered.</p><ol><li>Protection against unplanned start-up</li><li>Exhausting</li><li>Holding, blocking and stopping</li><li>Reversing the movement</li><li>Reducing speed</li></ol><h2>1. Protection Against Unplanned Start-Up</h2><p>One of the most important safety functions in a pneumatic system, or indeed any automated system, is to protect against the machine starting up accidentally. The most common way to do this is with a pneumatic safety valve which has been built for the purpose, and certified to the appropriate PLr. For PLc a simple solenoid valve is sufficient, with a single valve element which will cut in. More sophisticated valve constructions are required for PLd or e applications, which will often need redundant valve elements to assure full safety functionality. Some products may have inbuilt electronics for this function, while others may require a safety PLC or external safe relay to comply with requirements for safety certification.</p><h2>2. Exhaust or Air Pressure Dump Protection</h2><p>Dumping or exhausting of air pressure will probably occur at the same time as an unexpected start-up, and many modern safety components combine the two functions. A pressure supply valve will shut off the air supply in the event of any failure, preventing excess air pressure building up in the system. At the same time, air pressure within the system will be dumped safely, although a manual exhaust release can also be installed as a back-up. In the two-channel, three-valve system that is commonly used with higher PL applications, the downstream flow will be greater than the input supply, so that downstream pressure can still be reduced to a safe level, even if the input valve fails to shut off. More recent technology is likely to incorporate soft-start valves and position sensors to control unexpected start-ups and pressure dumps.</p><h2>3. Holding, Blocking and Stopping</h2><p>Holding, blocking or stopping operational equipment is one of the key safety functions in a pneumatic system, and applies to all kinds of applications across a wide PL range. It may be as simple as installing a pressure-operated check valve in the cylinder port. The trapped air pressure in the cylinder will hold the load and make sure it can&#39;t move, and a pressure switch added to the circuit could provide DC.</p><p>In more complex applications requiring higher PL, a mechanical lock or brake may be installed on the cylinder&#39;s piston rod. This clamps the piston rod into position using spring pressure and an appropriate valve. Another version of the rod lock comprises a pneumatic device installed on the piston rod. This device only allows cylinder movement if air pressure is directed to a port on the rod lock itself. Increased DC can be obtained by adding a sensor to detect when the rod lock is engaged.</p><h2>4. Reversing Cylinder Movement</h2><p>In simple, low PL applications, this safety function can be accomplished in a pneumatic system by a spring-return solenoid valve. Higher PL applications will require a more robust solution, such as a servo-pneumatic unit. The safety functionality in category 4 control architecture relies on the position of each valve element being sensed and transmitted to a safety PLC. This will ensure that the cylinder may only extend if both valve elements have moved.</p><h2>5. Reducing Speed</h2><p>Reducing speed is the easiest to implement of the key safety functions in a pneumatic system. It requires the installation of simple flow controls that can be validated with the appropriate movement parameters. Tamper-resistant locks may be installed to ensure these parameters are not altered.</p><p>This post was originally posted on&nbsp;<a data-fr-linked="true" href="https://www.rowse.co.uk/blog/post/5-key-safety-functions-in-pneumatic-systems" rel="noopener noreferrer" target="_blank">www.rowse.co.uk</a>.</p>

Machine building has become more complex with the introduction of more automated systems, and designers must devote more careful attention to the newer safety standards that apply.

5 Key Safety Functions in a Pneumatic System

The fourth industrial revolution and the technologies that create it explained: The whole is bigger than its parts.

Industry 4.0 Deep Dive. Part 1: Introducing smart manufacturing

<p>Automated systems governing smart manufacturing, and enabled by IoT devices, give us the IIoT (Industrial Internet of Things), which allows for the expansion and streamlining of all aspects of business. In addition, the deployment of AI and robots is particularly useful in industry as they increase product quality, production efficiency and overall speed.</p><p>The introduction of collaborative robots (or cobots) is becoming increasingly common, in industry as well as in laboratory work and commerce. Cobots can take instructions from humans in order to work more productively with them, including instructions that were not originally anticipated in the initial programming. Robots don&#39;t get bored, hungry or tired. The <a href="https://www.rowse.co.uk/blog/post/the-rise-of-artificial-intelligence" target="_blank" title="The Rise of Artificial Intelligence">capabilities of Artificial Intelligence</a> are far beyond human capacity when it comes to such things as miniaturisation and precision measurements, and they deliver vastly superior quality assurance.</p><p><img src="https://s3-us-west-1.amazonaws.com/wevolver-project-images/froala%2F1584368176061-1584368176061.png" style="width: 1080px;" class="fr-fic fr-dib"></p><p>So what are the benefits we can expect to see as industry moves forward into the 21st century world of cobots and AI?</p><h2>1. Direct Automation</h2><p>IIoT connects all IoT enabled devices to the factory floor, integrating manufacturing processes with big data and making them programmable via a logic controller. Increased use of precision sensory equipment means that information can be generated, recorded and analysed for all aspects of the production process, covering anything from temperature to item picking and packaging. Programmable logic controllers with AI capacity for deep learning can then respond automatically to the seamlessly generated information, and make alterations to the minutest function without recourse to human intervention. The big data analytics processed by AI can substantially improve performance across the entire production process, and can be operated remotely.</p><h2>2. 24/7 Production</h2><p>Human beings are biological organisms and need regular maintenance, e.g. food and sleep. For any production facility to continue working round the clock it is necessary to introduce shifts, using three human workers for every 24 hour period. Robots don&#39;t get tired or hungry, and are capable of working on the production line 24/7. This allows the expansion of production capabilities, which is increasingly necessary to meet the demands of worldwide customers. Furthermore, robots are more efficient in many areas such as the assembly line, and picking and packing departments. They can greatly reduce turn-round times in many areas of the business operation.</p><h2>3. Safety</h2><p>Human beings are fallible and prone to making mistakes, especially if they are tired or distracted. Errors and accidents do occur on the factory floor and in any construction or processing environment; a tendency which can be all but eradicated by AI and robotic assistance. Remote access control means a reduction in human resources, especially when the work is dangerous or requires superhuman effort. Even regular working environments will cut down on the incidence of industrial accidents and lead to an overall improvement in safety. In addition, more advanced sensory equipment integrated with IIoT devices make the installation of safety guards and barriers a simpler and more effective measure to protect human lives.</p><h2>4. Lower Operational Costs</h2><p>Many companies are viewing the introduction of AI into the manufacturing industry with trepidation, as it requires a huge capital investment. On the other hand, the ROI is significant and increases as time goes on. Once intelligent machines begin to take over the daily activities of a factory floor, businesses will benefit through considerably reduced operating costs, with predictive maintenance helping additionally to reduce machine downtime.</p><p>These days, consumers are increasing their demand for unique, personalised or customised products, while continuing to expect the best value. With such advances as 3D printing and IIoT connected devices, it is becoming simpler and cheaper to meet these needs, and using virtual or augmented reality design techniques means that the whole production process will be more cost-effective. Integrating machine learning and CAD means that systems can be designed and tested in a virtual model before they are put into production, thus reducing the cost of trial-and-error machine testing.</p><h2>5. Greater Efficiency</h2><p>IIoT enables the collection of vast amounts of data and advanced analytics which can be used to gain insights into consumer behaviour. Trends can be predicted, patterns recognised and market developments forecast across time, socioeconomic sectors and geographic markets, as well as taking into account political developments, macroeconomic cycles and even weather patterns. AI has the capacity with machine learning to anticipate information, to refine processes, and to track incongruities, all the way down the supply chain from source to finished product. This is particularly aided by such technologies as&nbsp;<a href="https://www.rowse.co.uk/blog/post/rfid-in-manufacturing" target="_blank" title="RFID In Manufacturing">RFID tracking</a>, which enable materials to be tracked without having to go through a physical process such as a bar code reader.</p><h2>6. Quality Control</h2><p>AI is also extremely useful for carrying out predictive maintenance on machinery and equipment. Using sensors to track performance and operating conditions, machines can learn to predict malfunctions and failures, and take action to remedy them before they occur. This can result in faster feedback, helping companies to eradicate unplanned downtimes.</p><p>Sensors can also detect the mot microscopic defects, scanning them at resolutions far beyond the capacity of human vision, thus improving productivity and increasing the percentage of items that will pass quality control. AI helps to speed up many routine processes and improves accuracy to an enormous extent. This obviates the requirement for quality control and in-process inspection by human beings, which is time-consuming and often fallible.</p><h2>7. Quick Decision Making</h2><p>When IIoT is coupled with cloud computing and virtual or augmented reality, companies can share simulations, confer on production activities and exchange critical or important information in real-time, irrespective of geographical location. The data gathered from sensors and beacons helps to determine consumer activity, allowing companies to anticipate future needs and make rapid decisions on production, as well as speeding up the exchange between manufacturers and suppliers.</p><hr><p>This post was originally posted on <a data-fr-linked="true" href="https://www.rowse.co.uk/blog/post/7-manufacturing-ai-benefits" rel="noopener noreferrer" target="_blank">www.rowse.co.uk</a>.</p>

As the world becomes increasingly automated, the IoT (Internet of Things) is already transforming our domestic and business lives. Nowhere is this more apparent than in the use of AI and robotics in the manufacturing industry, with all the benefits offered by Industry 4.0.

7 Benefits of AI In Manufacturing

<p class="lh li dz ap lj b lk ll lm ln lo lp lq lr ls lt lu" data-selectable-paragraph="" id="f2d4">Most robots used in factories today are powerful but blind. They are programmed to do specific tasks repeatedly with high precision and speed, but until now have not been able to adapt to changes in the environment or handling variables. Robot arms often lack the ability to grasp an object if it is moved a few inches, not to mention picking or sorting millions of different products in a warehouse.</p><p class="lh li dz ap lj b lk ll lm ln lo lp lq lr ls lt lu" data-selectable-paragraph="" id="1c10">As a result system integrators design shaker tables, feeders, conveyor belts, etc., so robots can be offered the same parts in the right orientation. Larger industrial robots need cages so they don’t accidentally hurt nearby people. Unlike human hands, which can perform tens of thousands of different tasks, robot arms can only be programmed for single tasks, and they need the proper end effectors.</p><p class="lh li dz ap lj b lk ll lm ln lo lp lq lr ls lt lu" data-selectable-paragraph="" id="43d5">Additionally, most robots cannot sense, unless you equip them with force or vision sensors. They don’t know if tasks are performed properly, because they cannot see or feel.</p><p class="lh li dz ap lj b lk ll lm ln lo lp lq lr ls lt lu" data-selectable-paragraph="" id="75cb">Traditional computer vision requires pre-registration of all objects, i.e. scanning and building computer-aided design (CAD) models for all items beforehand with expensive 3D cameras. The process is time-consuming and inflexible. Even if there was a box with exactly the same shape, but a different size, a robot couldn’t automatically recognize the differences and figure out how to grasp the new item.</p><p class="lh li dz ap lj b lk ll lm ln lo lp lq lr ls lt lu" data-selectable-paragraph="" id="85ad">In addition, translucent packaging, reflective surfaces, and deformable objects (such as food and clothing) make robotic vision an even more difficult problem to solve. The additional non-robot hardware and integration end up costing four times more than a robot arm, which costs anywhere between $10,000 to $100,000.</p><p class="lh li dz ap lj b lk ll lm ln lo lp lq lr ls lt lu" data-selectable-paragraph="" id="9cf8"><strong class="lj lv">Adding intelligence to robots</strong></p><p class="lh li dz ap lj b lk ll lm ln lo lp lq lr ls lt lu" data-selectable-paragraph="" id="41fb">To solve these issues, deep reinforcement learning (DRL) is unleashing the full potential of robotics, giving them the ability to recognize and react to their surroundings and handle product variations. Robots can now learn from experiences with off-the-shelf cameras and deep neural networks.</p><p class="lh li dz ap lj b lk ll lm ln lo lp lq lr ls lt lu" data-selectable-paragraph="" id="7081">With enough data and practice, robots can now teach themselves new abilities, as was seen with robots’ ability to identify images and win video games. With every grasp and trial, a robot gets smarter and more adept at mastering the task. Furthermore, cloud-connected robots allow learnings to be shared with each other. This is a tremendous shift, making robotics solutions more dexterous, flexible, and scalable.</p><p class="lh li dz ap lj b lk ll lm ln lo lp lq lr ls lt lu" data-selectable-paragraph="" id="0fae"><strong class="lj lv">Deep Learning</strong> (DL) is primarily used for image classification. Deep neural networks learn representations, rather than task-specific algorithms.</p><p class="lh li dz ap lj b lk ll lm ln lo lp lq lr ls lt lu" data-selectable-paragraph="" id="58a9"><strong class="lj lv">Reinforcement learning</strong> (RL) is inspired by behavioral psychology. It learns control policies by trial-and-error through reward or punishment.</p><p class="lh li dz ap lj b lk ll lm ln lo lp lq lr ls lt lu" data-selectable-paragraph="" id="ba32"><strong class="lj lv">Deep Reinforcement Learning</strong> (DRL) combines these two approaches. It is currently suited for:</p><p class="lh li dz ap lj b lk ll lm ln lo lp lq lr ls lt lu" data-selectable-paragraph="" id="d407">· Simple or simulated tasks</p><p class="lh li dz ap lj b lk ll lm ln lo lp lq lr ls lt lu" data-selectable-paragraph="" id="0cad">· Fault-tolerant tasks</p><p class="lh li dz ap lj b lk ll lm ln lo lp lq lr ls lt lu" data-selectable-paragraph="" id="bf6e">· Just enough variation to make explicit modeling difficult</p><p class="lh li dz ap lj b lk ll lm ln lo lp lq lr ls lt lu" data-selectable-paragraph="" id="b81f">· Easy-to-define goals (reward functions)</p><p class="lh li dz ap lj b lk ll lm ln lo lp lq lr ls lt lu" data-selectable-paragraph="" id="25e2">· Semi-structured environments</p><p class="lh li dz ap lj b lk ll lm ln lo lp lq lr ls lt lu" data-selectable-paragraph="" id="ca31"><strong class="lj lv">Warehouse as a testing ground for AI</strong></p><p class="lh li dz ap lj b lk ll lm ln lo lp lq lr ls lt lu" data-selectable-paragraph="" id="2bb1">Automation in warehouses is widely considered the “low-hanging fruit” for these technologies because tasks tend to be similar across warehouses, and the environment is more structured. For example, order picking represents more than 40% of operational costs across most warehouses. While tasks are repetitive, they involve some variation with products and packaging design, making it impossible to hard-program robots in advance.</p><p class="lh li dz ap lj b lk ll lm ln lo lp lq lr ls lt lu" data-selectable-paragraph="" id="bf53">Furthermore, labor costs account for up to 70% of a warehouse’s total budget. Pressured by e-commerce companies such as Amazon, retailers are looking to automation to address labor shortages and reduce costs. Customer demand, price competition, and fast delivery requirements are driving a strong business case for warehouses to be the first adopter of AI-enabled robots.</p><p class="lh li dz ap lj b lk ll lm ln lo lp lq lr ls lt lu" data-selectable-paragraph="" id="d7b2"><strong class="lj lv">Chart: Why do we see AI-enabled robots in the warehouse first?</strong></p><p class="lh li dz ap lj b lk ll lm ln lo lp lq lr ls lt lu" data-selectable-paragraph=""><strong class="lj lv"><img src="https://images.wevolver.com/eyJidWNrZXQiOiJ3ZXZvbHZlci1wcm9qZWN0LWltYWdlcyIsImtleSI6ImZyb2FsYS8xNTY4OTAzMzQ5MzI3LTFfTDhkY3A1c1lfcEU0MTdqdW9VaEMwZy5wbmciLCJlZGl0cyI6eyJyZXNpemUiOnsid2lkdGgiOjk1MCwiZml0IjoiY292ZXIifX19" style="width: 766px;" class="fr-fic fr-dib"></strong><br></p><p class="lh li dz ap lj b lk ll lm ln lo lp lq lr ls lt lu" data-selectable-paragraph="" id="76b9">Most tasks in a warehouse are fault tolerant, unlike tasks on a production floor or in the automotive industry. If a robot accidentally drops an object in a warehouse, it can try again as long as it achieves the required speed and success rates without damaging the goods.</p><p class="lh li dz ap lj b lk ll lm ln lo lp lq lr ls lt lu" data-selectable-paragraph="" id="d8c5">DRL is particularly suited for fault-tolerant simple tasks — actions a human can perform within a few seconds — with enough variation in semi-structured environments. As technology improves in environments like warehouses, it can then be applied in other industries, such as manufacturing, where tasks are more complicated and more uncertainties exist. Thus, an automated warehouse acts as an incubator, predicting a knowledge spill-over to many other market segments.</p><p class="lh li dz ap lj b lk ll lm ln lo lp lq lr ls lt lu" data-selectable-paragraph="" id="36d8"><strong class="lj lv">Piece-picking is the holy grail</strong></p><p class="lh li dz ap lj b lk ll lm ln lo lp lq lr ls lt lu" data-selectable-paragraph="" id="2047">Warehouse automation has been around for decades, in the form of automated storage, retrieval systems, and mobile robots like those from Kiva (now Amazon Robotics), that move items around.</p><p class="lh li dz ap lj b lk ll lm ln lo lp lq lr ls lt lu" data-selectable-paragraph="" id="f719">But most of the pick-and-pack operations within a warehouse are still performed by humans, with labor costs accounting for 50 to 70% of the overall warehouse budget. Worker shortages have driven up salaries by a rate of 6 to 8% every year, while robot costs have fallen considerably since the 1990s.</p><p class="lh li dz ap lj b lk ll lm ln lo lp lq lr ls lt lu" data-selectable-paragraph="" id="62e4">Piece-picking has long been the holy grail of robotics. Year after year, incumbents such as Amazon and KUKA host robotics challenges for startups and academic teams to create machines that can identify, pick up and stow goods. The emergence of deep learning allows robots to recognize, pick, and place hundreds or up to thousands of items.</p><p class="lh li dz ap lj b lk ll lm ln lo lp lq lr ls lt lu" data-selectable-paragraph="" id="e9b0">But the technology is not yet perfect. It’s still challenging for a machine to recognize tens of millions of objects, and manipulate deformable objects or items with transparent packages. However, our interviews with Locus Robotics and Osaro indicate that many industry experts expect the technology will mature. In addition, we will have a robot vision camera that can recognize almost everything in a warehouse within five years. If that happens, it will affect not just warehouses, but retail, delivery, and many other applications that we haven’t yet imagined.</p><p class="lh li dz ap lj b lk ll lm ln lo lp lq lr ls lt lu" data-selectable-paragraph="" id="9aae"><strong class="lj lv">Pushing towards standard models</strong></p><p class="lh li dz ap lj b lk ll lm ln lo lp lq lr ls lt lu" data-selectable-paragraph="" id="68ce">Technology improvements in warehouse automation and beyond will also push the mature, but fragmented, robotics industry to become more standardized and modular. Given the nature of narrow AI, most Silicon Valley startups in the AI-enabled robotics industry choose to focus on specific vertical applications and collect proprietary data as a defensive advantage.</p><p class="lh li dz ap lj b lk ll lm ln lo lp lq lr ls lt lu" data-selectable-paragraph="" id="155b">Piece-picking by robots is the best example. As the technology improves, we will see warehouse piece-picking startups enter manufacturing, food, and agriculture markets. Moving away from manually programming a robot and having it learn and act autonomously on specific simple tasks — and sharing the learning — will become more important in the future.</p><p class="lh li dz ap lj b lk ll lm ln lo lp lq lr ls lt lu" data-selectable-paragraph="" id="a46c">However, it’s not easy to find scalable and technically feasible business cases like piece picking that customers are willing to pay for. For example, assembly tasks vary from factory to factory, making it difficult to get sufficient training data or scale a solution across different customers.</p><p class="lh li dz ap lj b lk ll lm ln lo lp lq lr ls lt lu" data-selectable-paragraph="" id="b207">Given current technology constraints, there’s still a trade-off between precision and flexibility. Deep learning allows robots to handle variation, but these AI-enabled robots still cannot achieve the same precision and accuracy as traditional robots, at least not today.</p><p class="lh li dz ap lj b lk ll lm ln lo lp lq lr ls lt lu" data-selectable-paragraph="" id="d366">Some startups are providing modules or technology stacks for other players in the ecosystem to mitigate the complexity and scalability issues mentioned above. New areas where startups are likely to enter and disrupt include teleoperation modules, controllers that allow for safer human-robot collaboration, human-centered AI protocols, and sensors that can quantify human behavior.</p><p class="lh li dz ap lj b lk ll lm ln lo lp lq lr ls lt lu" data-selectable-paragraph="" id="c634">Eventually, we may see unified operating systems for AI-enabled robots similar to Windows or Android in the consumer electronics space, or standard user interfaces that will fundamentally transform the fragmented robotics industry and further accelerate its disruption.</p><p class="lh li dz ap lj b lk ll lm ln lo lp lq lr ls lt lu" data-selectable-paragraph="" id="6144">Standard modules and interfaces will bring down the cost to enter the market and shorten the time to market. As DRL and robotics technology improves, we will see a boom across industries including agriculture, food processing, home, and even surgical robotics.</p><p class="lh li dz ap lj b lk ll lm ln lo lp lq lr ls lt lu" data-selectable-paragraph="" id="2e46"><strong class="lj lv">How technology will transform the industry landscape</strong></p><p class="lh li dz ap lj b lk ll lm ln lo lp lq lr ls lt lu" data-selectable-paragraph="" id="6586">Led by industry leaders like Amazon and Ocado, a British online grocery store, retailers see efficient and cost-effective supply chains as a key to success and are incentivized to automate their warehouses for lower cost, lower management complexity, and higher scalability. While Amazon keeps its Kiva mobile robots and automated warehouse solutions for in-house use only, Ocado launched its fulfillment platform to help other companies store, pick, and transport groceries. But most retailers still rely on material-handling companies and system integrators to build and maintain an automated warehouse.</p><p class="lh li dz ap lj b lk ll lm ln lo lp lq lr ls lt lu" data-selectable-paragraph="" id="0f22">Incumbent material-handling companies tend to exploit the current technology, focusing on high-revenue-potential retrofit projects. This gives startups an opportunity to enter from the lower-end markets, where incumbents are less interested, and thus would not respond immediately. Dozens of startups offer scalable and affordable solutions that address customer needs for robots that can handle high-variability, low-precision tasks, such as piece picking in the warehouses.</p><p class="lh li dz ap lj b lk ll lm ln lo lp lq lr ls lt lu" data-selectable-paragraph="" id="5539">On the other hand, large robotics companies have long focused on developing high-speed and more-precise solutions to serve their main customers in the automotive and manufacturing industries. The pursuit of absolute precision is contrary to the process of deep learning, where results improve over time through trial and error.</p><p class="lh li dz ap lj b lk ll lm ln lo lp lq lr ls lt lu" data-selectable-paragraph="" id="7bbf">Some robotics companies saw an opportunity to enter new segments that require robots to handle high-variability, but lower-precision tasks, and began to build in-house AI teams and explore partnerships with startups developing deep neural network solutions for robots.</p><p class="lh li dz ap lj b lk ll lm ln lo lp lq lr ls lt lu" data-selectable-paragraph="" id="c818">There are currently more than 20 warehouse piece-picking robot startups worldwide, including Osaro and Righthand Robotics. Most of these startups (between seed round and Series C funding) are working on pilot projects with incumbent material-handling, retail, and robotics companies.</p><p class="lh li dz ap lj b lk ll lm ln lo lp lq lr ls lt lu" data-selectable-paragraph="" id="416d">Software and services will also capture more values, while premiums in hardware will decrease. With this power shift, as well as potential mergers and acquisitions, the landscape will change significantly over the next few years.</p><p class="lh li dz ap lj b lk ll lm ln lo lp lq lr ls lt lu" data-selectable-paragraph="" id="4459"><strong class="lj lv">The future of AI-enabled robotics</strong></p><p class="lh li dz ap lj b lk ll lm ln lo lp lq lr ls lt lu" data-selectable-paragraph="" id="4e2a">Thoughts vary on the effect that AI and DRL will have on the robotics industry. One scenario is the robotics industry will consolidate because of the AI network effect, which says results improve as the amount of training data they’re given increases. This idea is that whoever owns the most proprietary data will win the competition.</p><p class="lh li dz ap lj b lk ll lm ln lo lp lq lr ls lt lu" data-selectable-paragraph="" id="aef7">However, new researches in the areas of Transfer Learning, Meta Learning, and Zero-Shot Learning make it possible to learn from fewer and fewer examples, generalizing learning results across different fields. Cloud-based AI and robotics will also help accelerate learning and diffusion.</p><p class="lh li dz ap lj b lk ll lm ln lo lp lq lr ls lt lu" data-selectable-paragraph="" id="8b8a">As these technologies evolve, the landscape may face yet another significant change. This all starts from the warehouse!</p>

automation

ORGANIZATIONS.

SHAPING THE INDUSTRY.

Rowse

Photoneo

READY Robotics

Avestec Technologies

Adapta Robotics

Latest Posts