Luyện nói tiếng Anh bằng Shadowing qua video: Inside a warehouse run by AI robots

We're enabling the next generation of Amazon to really grow

⏸ Tạm dừng

Tốc độ:

149 câu

Nếu các câu quá ngắn hoặc quá dài, hãy bấm Edit để chỉnh sửa.

We're enabling the next generation of Amazon to really grow

and realize their full potential and driving the economy forward because this is the future of logistics.

I'm Andre, Chief Growth Officer at BrightPick,

and we're developing the next generation of warehouse robotics.

Our mission is really to enable any warehouse,

large or small, to fully automate their order fulfillment in a matter of weeks.

So right now we're standing in one of our installations

where we pretty much automated the entire fulfillment process using close to 50 BrightPick Autopicker robots.

I'm going to show you around so you can see how it looks.

So our flagship robot is called BrightPick Autopicker

and it's actually the only mobile robot in the world that not only transports items but can also robotically pick them.

And the big advantage with that is it actually does the pick directly in and out.

And we actually took a lot of inspiration from the way humans do picking because humans,

they go directly from one pick to the next and that is actually the most efficient way to pick orders.

Basically, what we did is you can think of BrightPick Autopicker

as almost like a human with a cart moving from one location in the warehouse to the next

and doing the picks as it goes on.

In fact, we're getting pretty close to human level performance and at a fraction of the cost.

So the way the robotic picking process works is the robots,

they retrieve a storage tote from the shelving,

they scan the storage tote,

create a 3D model of the tote.

Then we have an AI engine which tells the robot what item to pick based on probability of success,

and then the robot picks that item from the storage toad into the order toad.

The thing with robotic picking is that inevitably the robots will not be able to pick some items some of the time.

I think anyone who tells you they can pick all items 100% of the time are lying, frankly.

It's a big challenge in robotic picking in terms of what do you do as a fallback in that scenario.

We're actually not entirely relying on robotic picking here.

We incorporated a couple of goods-to-person stations into this installation,

both as a fallback in case the robots can't pick the items,

but also for items that aren't ideally suited for robotic picking.

So let's say items that are heavier or that have a changing center of gravity,

like liquids, for example.

So this is our goods to person station right here.

It actually works on the same principle as the robotic picking.

So each auto picker brings both the storage tote and the order tote to a human operator.

There's a tablet that tells him exactly how many items and

which items he needs to pick and then the operator picks it from one tote to the other tote.

So it's super simple, pretty ergonomic.

The robots they raise both totes to waist height.

They actually have four different sides.

So what that enables us to do is to feed the human operator from four sides at the same time.

So during peak hours when it's really busy,

there's basically four robots at the station at any given point in time,

which means there's like zero idle time in between fixed.

A really important part of our philosophy is product simplicity and cost.

You can see like we use a pretty simple SCAR arm on the robots.

It's two axes, the end effector is pretty flexible

and we could have of horse stuck like an expensive six-axis arm on the robot,

but that would have basically doubled the cost for an extra 10% of performance.

There's no point in doing that

when basically what we do in those 10% of the edge cases is we take those items to a human operator,

which you would anyways need because even if you have the most expensive robotic picker in the world,

there's still going to be some items when the robot can't pick them.

So you anyways have to solve for that.

And I think that's I think that's a really important part to understand about the whole BrightBix system is

it's designed with simplicity in mind.

So there's a lot of super complicated features,

AI, you know, the software orchestration behind it.

But the actual process of working with the robots and installing it in the warehouse,

it's super simple, right?

You see I'm like walking around with the robots.

Of course, usually we all want people to do that because it slows down the system.

But it's perfectly safe, you know,

you don't need any safety fencing.

There's no QR codes on the floor for navigation.

We use LiDARs and AI.

The shelving, it's like very cheap.

It's basically the same shelving you can buy in Home Depot,

which also means like you can install in a matter of weeks and even move the shelving from one warehouse to another,

which, you know, it's pretty important.

You don't want to build like a monument in your warehouse that you're stuck with for the next 10 years.

That's actually the reason we purposefully don't use any conveyors because,

you know, once you build a conveyor good luck moving that.

That's basically why we developed the system is to enable any retailer to automate their fulfillment in a matter of weeks.

This is really like a decade of development behind this.

I mean first we developed the 3D scanner which is actually from our sister company Photo Neo.

We then developed the AI picking engine.

I think we've done close to a billion picks by now to train our AI.

We then developed a mobile platform,

all the software orchestration that goes with that.

We actually had a goods-to-person solution before this,

but then ultimately we brought it all together inside the Autopicker.

And now this robot can do everything you need for order fulfilling, right?

Like robotic picking, goods-to-person picking,

pallet picking, order buffering, consolidation, dispatch, replenishment.

When we think of BrightPick Autopicker as basically a humanoid in all but name,

If you think about what a human picker does in a warehouse,

it's basically the same, except auto pickers are able to do it much faster, more efficient.

It's literally a tenth of the cost of a humanoid.

We don't believe in technology for technology's sake.

Why do you need a human form factor when you can basically achieve human level performance at a fraction of the cost?

So a common misconception that people have about robotics is that we're taking away people's jobs,

but the reality is actually completely opposite because pretty much every customer that buys our solution,

they're dealing with massive labor shortages,

whether it's in logistics or in manufacturing,

and there's simply not enough people to meet demand because we have a shrinking labor force.

These are physically tough jobs.

Imagine walking two marathons a week just to pick orders,

100

and there's not enough people to fill the demand, basically.

101

For example, this customer right here,

102

they had to actively turn off their marketing spend because they didn't have enough people in their warehouse to pick orders.

103

And now after they installed our solution,

104

they're able to grow it 100%.

105

It was basically like three people servicing our entire system.

106

Whereas before they would have needed 20, 30 people here.

107

Another feature that's unique to BrightPick is what we call pallet picking.

108

And this is actually, imagine you have items that are fast movers.

109

So let's say you're putting in samples or you're basically selling products that go into every order or every second order.

110

It doesn't really make sense to put those items into the system

111

because you're going to be replenishing them almost as fast as you're taking them out.

112

So what we developed, what we call these pallet picking positions,

113

where basically human operators are able to pick directly from pallets or from cases,

114

which you see stored behind them,

115

and place them onto the order tote at the robot.

116

So that basically makes the replenishment process super easy,

117

where all you do is you stick a pallet there or you throw a case there,

118

and then the robot pulls up,

119

tells the human operator on the tablet what item to pick,

120

and the human can pick that item and place it onto the robot.

121

So you really get that flexibility of robotic picking,

122

GTP picking, and pallet picking,

123

all as part of one system.

124

You really don't need any other picking.

125

So this right here is the buffer area,

126

and actually this is a 24-7 operation,

127

so the robots are working even at night when there's practically no one in the warehouse.

128

and what they're doing is after they pick the orders,

129

they buffer them in this area.

130

And then once the morning shift comes in,

131

they take them out, deliver them to pack out to be packed and shipped.

132

Another use case is imagine you have an order that consists of 10 picks,

133

but only nine of them are robotically pickable.

134

So the robots are gonna pick those nine items overnight,

135

buffer them, and then when the morning shift comes in,

136

it's gonna do that last 10th pick using the goods to person or pallet station.

137

And basically then the order is complete.

138

So because we have such a high level of automation,

139

we're able to easily operate at night because you practically don't need anyone to run the system.

140

We're now in a hyper-growth mode.

141

We're really focused on ramping up our manufacturing because we already get more demand than we can meet for our solution.

142

So now it's really about laying down the groundwork to make

143

sure we can manufacture as many of these robots as possible and really ensure that they're as reliable as possible.

144

We understand that these are mission critical processes,

145

so we make sure that each robot can really operate five,

146

10 years without breaking down.

147

This is one of many installations that we've already done.

148

We have many more in the pipeline,

149

so right now it's rolling up our sleeves and getting to work.

TRENDING

Phổ biến

Phương Pháp Shadowing Là Gì?

Shadowing là kỹ thuật học ngôn ngữ có cơ sở khoa học, ban đầu được phát triển cho chương trình đào tạo phiên dịch viên chuyên nghiệp và được phổ biến rộng rãi bởi nhà đa ngôn ngữ học Dr. Alexander Arguelles. Nguyên lý cốt lõi đơn giản nhưng cực kỳ hiệu quả: bạn nghe tiếng Anh của người bản xứ và lặp lại to ngay lập tức — như một "cái bóng" (shadow) đuổi theo người nói với độ trễ chỉ 1–2 giây. Khác với luyện ngữ pháp hay học từ vựng bị động, Shadowing buộc não bộ và cơ miệng phải đồng thời xử lý và tái tạo ngôn ngữ thực tế. Các nghiên cứu khoa học xác nhận phương pháp này cải thiện đáng kể phát âm, ngữ điệu, nhịp điệu, nối âm, kỹ năng nghe và độ lưu loát khi nói — đặc biệt hiệu quả cho người luyện IELTS Speaking và muốn giao tiếp tiếng Anh tự nhiên như người bản ngữ.

ShadowingEnglish.com – Luyện Nói Tiếng Anh Bằng Shadowing

Nói tiếng Anh lưu loát với phương pháp Shadowing — nghe người bản xứ, đọc theo từng câu, rèn phản xạ ngôn ngữ thật sự. Được hàng ngàn người học IELTS tin dùng.

Luyện nói tiếng Anh bằng Shadowing qua video: Inside a warehouse run by AI robots

Phổ biến

Ngữ Cảnh & Nền Tảng

5 Cụm Từ Hữu Ích Cho Giao Tiếp Hàng Ngày

Hướng Dẫn Shadowing Một Cách Từng Bước

Phương Pháp Shadowing Là Gì?

Luyện nói tiếng Anh bằng Shadowing qua video: Inside a warehouse run by AI robots

Tải Ứng Dụng

Ngữ Cảnh & Nền Tảng

5 Cụm Từ Hữu Ích Cho Giao Tiếp Hàng Ngày

Hướng Dẫn Shadowing Một Cách Từng Bước

Phương Pháp Shadowing Là Gì?