A $10 mesh robot moves the classroom bottleneck from the purchase order to the radio

Forty percent of the robot is the chip and about thirty-five percent is the drivetrain, since the two geared motors and the L298N driver together come to $3.50 ^[15][3]. The radio that makes coordination possible costs roughly what the wheels cost, which is why the choice of an ESP32 over an Arduino or a Raspberry Pi is doing the real work: LearnOBots reports the mesh stack is native to ESP-IDF, so there are no shields, no add-on modules and no wires between robots ^[7].

The sub-dollar lines are where a school actually pays. The 18650 cell is $0.90 because it is recycled from a laptop pack, the motors are $2.20 because they are surplus from old toys or local vendors, and the chassis is $0.75 because it is laser-cut, printed, or cardboard ^[11]. For one robot that is a weekend of scrounging. For thirty it is thirty used lithium cells that somebody has to source, match and check before a room of children charges them. None of that appears on an invoice, and all of it lands on a teacher or a lab technician.

The classroom arithmetic is where the cost curve actually bites. Under the shared-kit pattern the post describes, a class of thirty works in teams of four or five, which is six robots in the room ^[6][19][17]. One node per child is thirty, a five-fold increase in nodes for the same hour and the same floor space ^[17]. Six robots can demonstrate a swarm. Thirty can be partitioned, starved of relays, and made to reform, which is the subject the post says it wants to teach: managed flooding, message propagation, and what happens when the node in the middle drops out ^[8].

Provenance deserves a line. This is one practitioner writing up his own classroom results, with parts priced in Pakistani rupees from local markets, and the reliability figure is self-measured rather than replicated by anyone else ^[2][3]. The build reads as tested rather than theoretical, which is worth more than most vendor documentation, but the ceiling it reports is the number the author reached, not a limit anyone has independently probed ^[8].