Basketball Robot: From Peach Baskets to ABB Arms

How the basketball robot grew from peach baskets and shooting machines to Toyota's CUE and humanoid players in 2026, and where ABB robot arms fit into sport.

INDUSTRIAL ROBOTICS

Chat With Robot

10/8/20265 min read

A basketball robot can mean three different things: a machine that feeds you passes in the gym, a robot that shoots with perfect form, or a humanoid that dribbles and plays. All three are moving fast in 2026, with China announcing an international humanoid basketball tournament for next spring. This post traces the basketball robot from the peach baskets of 1891 through shooting machines and Toyota's record setting CUE, then looks at humanoid robot sports today, where ABB's robot arms come into the story, and what a home or school court might get in the next few years.

Black shooting machine with a tall net frame standing under a basketball hoop in a blue gym
Black shooting machine with a tall net frame standing under a basketball hoop in a blue gym

A Gun basketball shooting machine under a hoop in the University of Kentucky practice gym. Photo: Basketballshootingmachine / Wikimedia Commons (CC BY-SA 4.0)

How basketball practice went from peach baskets to shooting machines

Basketball began in December 1891, when James Naismith, a teacher at the YMCA training school in Springfield, Massachusetts, nailed two peach baskets to a gym balcony and gave his class a soccer ball. The baskets still had their bottoms, so after every score someone had to climb up and get the ball out. Open nets came later, and the game spread quickly through colleges. By the mid 1890s universities such as Syracuse and Yale had teams and published their photographs in official guides.

Old black and white photo of a man in a suit holding a ball and a woven basket
Old black and white photo of a man in a suit holding a ball and a woven basket

James Naismith, who invented basketball in Springfield in 1891, holding a ball and a peach basket. Photo: unknown author via Wikimedia Commons (public domain)

For most of the twentieth century, shooting practice depended on a second person. A coach, teammate or parent stood under the hoop, chased rebounds and passed the ball back. Players who wanted a few hundred shots a day needed a lot of patient helpers. The first help came from simple nets and ramps that steered made baskets back toward the shooter. They worked for free throws but did little for misses that bounced off at odd angles, which is most of what a young player shoots.

Seven young men in numbered vests, one holding a ball, in an old printed team photo
Seven young men in numbered vests, one holding a ball, in an old printed team photo

The Syracuse University basketball team in the Official A.A.U. Basketball Guide of 1894. Photo: Internet Archive Book Images via Wikimedia Commons (no known restrictions)

Shooting machines solved that. A machine like The Gun catches made and missed shots in a large net above the rim and fires the ball back with a passing wheel, rotating to follow the shooter around the arc. College and pro teams use them in their practice gyms, and newer versions track every shot and change the passing pattern. At CES in January 2025, a startup went further and showed a robot that uses AI to chase down rebounds on its own. Every one of them still needs a human to take the shot.

A man in a blue shirt holds a basketball next to a shooting machine on a court
A man in a blue shirt holds a basketball next to a shooting machine on a court

Duke coach Jon Scheyer beside The Gun 14X Smart Pass shooting machine, 2025. Photo: Basketballshootingmachine / Wikimedia Commons (CC BY-SA 4.0)

Robot shooters and humanoid players in 2026

The best known robot shooter is Toyota's CUE, built by volunteer engineers who took their inspiration from the basketball manga Slam Dunk and first showed it in 2018. In 2019 CUE3 set a Guinness World Record by making 2,020 consecutive free throws, a task that took about seven hours. The sixth version, CUE6, then set a record for the farthest shot by a humanoid robot, sinking the ball from 24.55 metres in Nagakute, Japan, nearly the full length of a court. CUE is listed as a member of the Japanese pro team Alvark Tokyo.

Two small black humanoid robots walk on artificial grass near a centre circle
Two small black humanoid robots walk on artificial grass near a centre circle

Humanoid robots in the Kid Size League at RoboCup 2016 in Leipzig, a robot sport that started long before robot basketball. Photo: ubahnverleih / Wikimedia Commons (CC0)

This year the focus has moved to humanoids that move like players. On 29 September 2026 the Guangdong provincial government said it will hold an International Humanoid Robot Basketball Challenge in Guangzhou in April 2027. Competition robots must have a head, torso, two arms and two legs, walk upright and get up on their own after a fall. They will compete in free throws, moving three point shots, passing with a human partner, dribbling endurance and a dribble and layup course, and every robot must carry an emergency stop. A freestyle show event is open to any design.

Grey humanoid robot with a glowing face plate standing on a patterned carpet
Grey humanoid robot with a glowing face plate standing on a patterned carpet

A Unitree G1 humanoid at an expo in 2024, a robot of the two legged, two armed kind the Guangzhou rules ask for. Photo: Sayanesy / Wikimedia Commons (CC0)

ABB does not build basketball robots, and its arms are bolted to the floor, so they will never dribble. They have still played games. Google's GoalsEye table tennis research used an ABB IRB 120, and a YuMi played tic tac toe with visitors at the 2019 FIRST Championship in Detroit. A fixed arm like the IRB 1300 could, in principle, run a repeatable shooting test rig for ball or hoop makers, and a GoFa cobot could hand balls to a player because it stops when it senses contact. Our table tennis robot post covers the arm side of robot sport in more detail.

Two armed ABB YuMi robot moving game pieces on a lit tic tac toe board
Two armed ABB YuMi robot moving game pieces on a lit tic tac toe board

An ABB YuMi playing tic tac toe with visitors at the FIRST Championship in Detroit, 2019. Photo: Stilfehler / Wikimedia Commons (CC BY-SA 4.0)

Where basketball robots go next, at home and in the factory

For homes and schools, the next step is cheaper rebounders that see the player. Cameras and AI shot tracking are already common in team gyms, and a driveway machine that counts shots, reads form and passes to the right spot is an easy extension. Humanoid players are much further off. A robot that can dribble, avoid a defender and finish at the rim has to balance, see and plan at once, and the Guangzhou events, from free throws to the layup course, test those skills one at a time.

Two white adult size humanoid robots line up to kick a soccer ball
Two white adult size humanoid robots line up to kick a soccer ball

NimbRo-OP2X humanoid robots from the University of Bonn playing soccer at RoboCup 2018 in Montreal. Photo: Sven Behnke / Wikimedia Commons (CC BY-SA 4.0)

Factories are part of this as well. Humanoids and ball machines need motors, gears, hands and sensors in large numbers, and companies such as Unitree already sell humanoids like the G1. Assembly lines for those parts use the same industrial robots found in car and electronics plants, including ABB arms programmed in RobotStudio. A sports robot boom would mostly show up as more orders for that kind of automation, though no ABB contract with a sports robot maker has been announced.

Two students holding a robot controller next to a small orange ABB robot arm on a workbench
Two students holding a robot controller next to a small orange ABB robot arm on a workbench

Students program an ABB IRB 120, the robot family Google used for its GoalsEye table tennis player. Photo: Peterantala / Wikimedia Commons (CC BY-SA 3.0)

Cost comes first. Team shooting machines are expensive and a capable humanoid costs far more, so most players will meet these robots at a school or club rather than at home. A falling humanoid or a fast pass can hurt someone, which is why the Guangzhou rules demand emergency stops. Coaches also have to learn what to do with all the shot data, since a machine can count makes and misses but cannot yet explain to a player why the elbow drifts.

A worker guides a white ABB GoFa cobot arm by hand at a work station
A worker guides a white ABB GoFa cobot arm by hand at a work station

ABB's GoFa cobot is guided by hand and stops when it senses contact. Photo: ABB

Innovation

AI solutions for effortless ABB robot control.

Automation

Robotics

ceojohntran@chatwithrobot.net

+84905311611

© 2025. All rights reserved.

qtran1215@gmail.com