blog posts

The Evolution of Robotics

The Evolution of Robotics

This article reports on some examples of the latest advances in industrial robotics, which show that there are many motivations for the rapid development of robotic.

روباتیک، همچنان در حال پیشرفت

Image: Self-propelled drone

Nothing can hold this drone down. Uses elytra, a collection of beetle-inspired wings, to correct itself

When your life falls apart, you have to get up again. Ladybugs take this advice seriously in the true sense of the word. If the insects get stuck in the back, they can use their hard outer wings, called elytra (recently known as Minecraft), to correct their position in a fraction of a second.

Inspired by this approach, researchers have synthesized self-propelled drones with Ultra. Simulations and experiments show that the artificial elytra can help rescue fixed-wing drones from dangerous situations and improve the aerodynamics of vehicles during flight.

Beetles, including ladybugs, have been around for tens of millions of years. They have developed several survival mechanisms that inspire modern robotic applications during this time.

The construction team was particularly fascinated by the beetle’s elite robotics, which are the outer wings of their famous black and red spots for ladybugs. Below the elytra in the hind wing is a translucent appendage used for flight.

When the ladybugs are on their backs, they use their elytra to stabilize themselves and press their hind legs or wings back and forth. The team designed Micro Aerial Vehicles (MAVs) that use the same technique but have actuators that modify their force. “Like the insect, the artificial elite has a degree of freedom that allows it to change direction if the vehicle overturns or overturns,” explains Wortsis.

The researchers of robotic created and tested artificial ultrasounds with different lengths (11, 14, and 17 cm). And different torques to determine the most effective combination for straightening a fixed-wing drone. While torque had little effect on performance, elytra length did.

On a smooth, hard surface, shorter elytra lengths produced different results. However, longer lengths were associated with a higher success rate. The longer elytra were then tested in different slopes of 10 °, 20 °, and 30 ° and different directions. The drones used the elytra to correct themselves in all scenarios, except one position on a steep slope. The design was also tested on seven soils: paving, sand, fine sand, stone, bark, wood chips, and grass. The drones were able to improve at full success rates in all areas except grass and fine sand.

Robotic robots want to move boxes with chopsticks. This approach can be more versatile than clamping or suction.

A pair of white robots with a black tube attached to each. The tubes are pressed against the sides of the cardboard box to lift it.

روباتیک، همچنان در حال پیشرفت

Image: Skillful robotics

Aside from the hustle and bustle, there are not necessarily many areas where robots have the potential to enter an existing workflow and instantly deliver a significant amount of value. But one area we’ve seen several robotic companies enter recently is box manipulation. In particular, the use of robots to unload boxes from the back of trucks, ideally much faster than humans. This is a good thing for robots because it works on their strengths: you can work semi-structured. And the usually predictable environment, speed, power, and accuracy are all valuable. And this is not a job that humans are particularly interested in. Have or are designed for them.

In the recently proposed approach, the box is manipulated without any suction or, in fact, any clamps. Instead, they use what might be best described as a pair of moving arms, each holding a robotic chopstick.

We can choose anything using chopsticks for robotic . If you are good with chopsticks, you can pick single grains of rice and pick things that are relatively large compared to the chopsticks scale. You think about limitations, so isn’t it interesting to have a robot that can manipulate things with a chopstick?

Interesting, but are there any practical reasons why using chopsticks to manipulate the box is a good idea? Of course, there is! The good thing about chopsticks is that they can handle almost anything (even if you enlarge them) and are especially valuable in confined spaces where you have a lot of differences in shape, size, and weight. They are also good for manipulation and can move and relocate objects accurately. And while the focus is initially on a trailer unloading task, having this additional manipulation capability allows for more difficult manipulation tasks in the future. Such as loading a trailer, a task that necessarily takes place as much as unloading. Fall is much more complicated.

Even if the chopping technique has obvious advantages, it also has disadvantages. And the biggest one is that it is much more difficult to use a manipulation technique like this. “The downside of chopsticks is that every human being tells you you need sophisticated control software to work. “But that’s part of what we’re playing: not just a clever hardware design, but software to work with.”

Meanwhile, what we have seen so far from other companies in this space is the completely consistent use of suction systems to move the box. If you have a smooth, impenetrable surface (like most boxes), suction can work quickly and reliably with minimal fancy planning.

Suction, however, has limitations for manipulation because it is inherently very sticky, meaning that doing anything accurately can be difficult and time-consuming. Other suction issues include its sensitivity to temperature and humidity its tendency to swallow as much dirt as possible. And the fact that you have to design the suction array based on the largest and heaviest things you anticipate. Confront. This last one is a special problem because if you want to manipulate smaller objects as well, you try with a suction array that is much larger than what you want.

Ideally, you want to make something that works the first time, not have to repeat it over and over again. Getting it right on the first try is probably unrealistic, but the better you can simulate things in advance, the closer you get to it.

“What we were able to do was set up our entire programming perception for robotic. And control system to look exactly like when that code was executed on a real robotic robot. When we run something on a simulated robot, it is 95% true, which is unprecedented.

A real-time simulator and software can be transmitted directly between the SIM card and the real one using high-fidelity hardware modeling. It can skillfully model how its system works, even on very complex simulation objects such as contact and adhesion. . The idea is that the result will be a system that can develop faster and at the same time perform more complex tasks better than other solutions.

 

روباتیک، همچنان در حال پیشرفت

Parrot unveiled a bug-inspired 4G drone. ANAFI Ai is designed for professionals and has a communication board (almost unlimited).

Image: A white quadcopter with an onion front and a gold camera mounted on a gray gimbal. Parrot drones.

Parrot launched the Anafi

robotic drone almost four years ago. Many are fans of small-consumption drones – beautiful design, great portability, great camera, and easy flight. But the biggest problem with Anafi (especially four years later) is that it does not look like the smartest drones, without any of the obstacles that are now standard on airplanes. Today, Parrot introduces Anafi AI, a professional redesign of Anafi that adds obstacle avoidance to a huge camera. And a 4G connection that allows the drone to fly anywhere (and behind any object) wherever you can. A reliable 4G signal.

While the design of the Anafi AI is somewhat reminiscent of the previous Anafi (probably because Parrot says both designs are insect-inspired), the look is quite different when you run out of battery. The UAV “head” is around what appears to be a massive heatsink (with a dedicated fan at the top). Along with a completely new camera system built into the front. The main camera (center) sits on a three-axis gimbal with mechanical and electronic image stabilization and has a 48-megapixel sensor that can record 4K video at 60 frames per second.

Obstacle cameras are located on each side of the main camera.

The press release said that the obstacle avoidance system “detects obstacles in all directions,” the press release said. Thanks to its amazing 311-degree rotation ability independent of the main camera, you were doing so with the two cameras. The idea is that if you are flying up, down, or back, the stereo system will rotate around to avoid obstacles in that direction. Not all directions, not all at once, but it is an innovative way to reduce mass, cost, and computational overhead.

The unique feature of Anafi AI is its 4G connection (with its SIM card), according to a press release.

With 4G, the data link between the drone and the pilot becomes stronger in any situation. 4G, already widely and reliably deployed worldwide, offers long-range transmission over low-frequency bands in the 700MHz-900MHz band. In addition, 4G connectivity means that professionals can work with UAVs at any distance.

There are many places where 4G is bad or non-existent. Fortunately, 4G connectivity is optional and not required, and you can still connect directly to the drone via the phone and radio on the controller. Switching between the radio controller and 4G is seamless. And with the controller as a recursive option, we can certainly get the (potential) benefits of 4G. However, it is important to note that drones flying beyond the line of sight are generally not permitted by the FAA here in the United States. We asked Parrot CEO Henry Seedoux this question because there seems to be a significant limitation to one of Anafi’s most distinctive AI features:
We believe in flying out of sight, and we believe it matters, and we believe that 4G is the safest radio system you can have for this. Many countries have an incentive and purpose of allowing drones to fly out of sight. These regulations are coming from country to country step by step. And we believe that this is the future of UAVs.

The last thing to note about Anafi AI at this point is that Parrot seems to be focusing heavily on making the drone more accessible to developers.

The core of the company’s new FreeFlight 7 software is open source. And there is an app development kit for both iOS and Android apps. Most importantly, you can write your custom code that can be executed while flying on a drone. Six provides an example of a wind turbine inspection – a developer can write code to fully automate this for wind turbines of any size and location. As a developer, you can access everything through Anafi SKD, including flight sensors (plus obstacle avoidance sensors), da dynamic networks, and even Internet access.

All of this sounds promising, but of course, we only have images and some optimistic marketing material at the moment. Like any other robot, the real test is to see how well Anafi AI works in the real world.

ُخعقزث:https://rasekhoon.net/article/show/1601015/%D8%B1%D9%88%D8%A8%D8%A7%D8%AA%DB%8C%DA%A9-%D9%87%D9%85%DA%86%D9%86%D8%A7%D9%86-%D8%AF%D8%B1-%D8%AD%D8%A7%D9%84-%D9%BE%DB%8C%D8%B4%D8%B1%D9%81%D8%AA