FEATURE Automated warehousing
have found their niche in the dynamic environment of warehouses, particularly those that have been slow to automate and have challenging pathways. The intricate mechanics of artifi cial ligaments and knees in the most advanced bipedal robots allow them to gracefully navigate uneven surfaces, climb stairs and ladders, and go through narrow spaces. If they were to stumble, they act just like a human would – catching themselves and going right back up. Their unique ability to step over, jump or manoeuver around objects off ers another distinct advantage over wheeled robots. Although bipedal robots use same type sensor technologies as their wheeled cousins, there are many more of them. For example, robot every joint – fi nger, elbow, toe, knee, hip, and so on – needs its own angle detection sensor. Bipedal robots need ultra-low latency, very high performance and highly- accurate types of sensors. For bipedal robot stabilisation, sensor accuracy and performance are key. In previous generations they were susceptible to damage from falls if their balance was off when walking, especially whilst carrying awkward loads. By integrating 6-axis IMUs, which combine a 3-axis gyroscope and 3-axis accelerometer, real-time roll, pitch and yaw measurements can help compensate for any instability. These IMUs must be very sensitive. To synchronise multiple bipedal robots,
there are additional compute power requirements and greater complexity to building the ML models; it can be done, but it is time-consuming and expensive. Another current limitation to the rise of bipedal robots is the overall speed at which they move, which is slower than an AMR moving in a straight line at about 3m/s, and with weight limitations for payloads. As with collaborative AMRs, these can also go into self-charge mode when they’re ready.
Drones: air robots? Drones are also increasingly important in warehouse and industrial settings. For some companies the primary uses for drones are inventory checks, indoor surveillance, maintenance checks, security, and more. Drones are very accurate and avoid obstacles, similarly to AMRs and bipedal robots. Most of the technology used in drones is the same as in AMRs and bipedal robots. However, the more sophisticated
automationmagazine.co.uk
drones integrate a pressure sensor to supplement the IMU, which improves the accuracy of altitude hold. Also, because the air pressure can vary across a facility, a second reference pressure reading from a grounded AMR can be more accurate in a swarm autonomy setup. Furthermore, additional ultrasonic ToF sensors can be used on the top of the drone to help it avoid hitting the ceiling, and on its base to enable smoother and safer landings, particularly when approaching a recharging cradle. There are regulatory constraints for the use of drones in indoor settings. Many countries require an operator with an appropriate pilot certifi cation or license, and be within line-of-sight of the drone at all times, to ensure safe operation. There may also be restrictions on the size and weight of the drones to minimise the risk of injury in case of collisions or accidents. Also, some countries implement height restrictions. Most drone regulations are subject to revisions as technology advances.
Solutions for developers Whether the robot is industrial, wheeled, bipedal or a drone, they often rely on similar technologies – IMUs, magnetometers, pressure sensors, microphones, ultrasonic ToF sensors, embedded motor controllers, temperature sensors, tilt and angle sensors, as well as advanced AI and ML software. In building these type robots, developers often face a signifi cant challenge – the need to source various technologies from multiple vendors. Typically each sensor technology
requires its own dedicated board, which typically comes from diff erent suppliers. Complicating matters further, these sensor boards must be compatible with various processor boards. Adding to the complexity is the presence of numerous open-source requirements from ROS- Industrial, including both ROS1 and ROS2.
Recognising these frustrations, TDK sought to simplify and expedite the development process by introducing TDK RoboKit 1, a streamlined board and reference design that allows to evaluate and develop robotic projects with minimal reliance on multiple vendors, accelerating development and reducing time-to-market.
Realising the potential As robots become more intelligent, gaining awareness of their surroundings and understanding objectives, there is a potential for them to operate with a greater degree of independence. Theoretically, it should be possible to have multiple robots of diff erent types collaborating seamlessly in the same warehouse or industrial park, each contributing to overall productivity and effi ciency. Such solutions can remove the most dangerous or monotonous tasks from humans, improving overall health and safety.
Whilst the practical implementation of this vision also poses challenges to making it an immediate reality, the innovations of technology companies and developers will continue to move us closer to a world of new industrial automation potential.
Automation | March 2024 33
Page 1 |
Page 2 |
Page 3 |
Page 4 |
Page 5 |
Page 6 |
Page 7 |
Page 8 |
Page 9 |
Page 10 |
Page 11 |
Page 12 |
Page 13 |
Page 14 |
Page 15 |
Page 16 |
Page 17 |
Page 18 |
Page 19 |
Page 20 |
Page 21 |
Page 22 |
Page 23 |
Page 24 |
Page 25 |
Page 26 |
Page 27 |
Page 28 |
Page 29 |
Page 30 |
Page 31 |
Page 32 |
Page 33 |
Page 34 |
Page 35 |
Page 36 |
Page 37 |
Page 38 |
Page 39 |
Page 40 |
Page 41 |
Page 42
