It is hard!

Many experts are needed!

Part of the problem is the nature of robotics. It is not a singular field; it involves combining both hardware and software with a deep and clear knowledge of business processes and physical workflow to build a robot that will actually function and add value. In order for a robotics company to craft a robot with autonomous capabilities it needs to employ highly accomplished people with expert skills in advanced mathematics, physics, mechanical and electrical engineering, complex programming and other supporting technical fields. These disciplines must also work in concert to create an autonomous robot that can work simply, safely, effectively and at a reasonable cost.

Moravec’s Paradox : Simple is Hard

Moravec's Paradox is a principle in artificial intelligence and robotics. It states that reason (conscious, intelligent, rational thought) is easy for machines to imitate, but that unconscious sensorimotor skills and instincts are, relatively, far more difficult. This contradicts the tradition in Western thought that consciousness and reason are the "highest" and "noblest" human faculties. The principle was first articulated by Hans Moravec, Seegrid co-founder.

Perception and Reality

Another challenge for robotics companies is the image people have of robots, mostly created by science fiction writers and Hollywood. There is a gap between what people believe robots can do and what they are capable of doing. It is accepted today that robots are building cars but in reality robots are just doing the same five welds over and over, and if you put all the pieces of a car on the floor no robot today could assemble it. In fact, if you moved the part it is supposed to weld just 6 inches away the robot won’t be able to do it.

For the most part robots do simply defined activities in distinct spaces with limited variability like spot welding a car frame or inspecting chips for damage. Up until very recently they have not been able to do useful work in unstructured environments such as warehouses and distribution centers.

The most effective robots need to Sense, Move, Analyze, Interact and Repeat – to truly become autonomous.

Sense. Move. Analyze. Interact. Repeat.

Sense - Humans have five senses that work in concert to provide a full and accurate picture of an environment. Robots have no such mechanism to fuse disparate data and draw conclusions about it. The first step is getting a good feel for what the environment looks like; if the environment is always the same this is less of a problem but when traveling through uncharted territory good sensory analysis capability is essential.

Move - Once the robot has an accurate characterization of its environment it can begin to determine its navigational course. Surprisingly this is not as simple as it would seem. The robot needs to know its exact location in its relationship to the current environment before it can start to move. The robot must be able to continually validate its location to the environment as it continues to move.

Analyze - Once a robot begins to move, it becomes advantageous to identify objects and their context, again harder than it seems. For example how does a robot know that a cup is not part of a table or that the elephant is far away and not just really small?

Interact - The robot now has an accurate characterization of the environment, knows its exact location, has the ability to move and navigate, can determine and identify objects, interfaces with a human or machine to accept a work order; therefore, it can now act and perform its assigned task.

Repeat - To be truly effective a robot must be capable of exactly repeating the assigned path and task repetitively without human intervention or correction to achieve true ROI (Return on Investment) efficiency.

Autonomous mobile robots, which were considered wildly impossible in the 1970s and 1980s, became experimental demonstrations in the 1990s: mobile robots mapped and navigated unfamiliar office suites, robot vehicles drove themselves, mostly unaided, across entire countries, computer vision systems located textured objects and tracked and analyzed faces in real time. Programs that recognized text and speech became commercially successful. Market success extended to physical robots as the 2000s began: Sony has sold hundreds of thousands of the AIBO robot pet despite its $1,000 plus price tag, and several small robot vacuum cleaners, especially the affordable iRobot Roomba, seem to be gaining consumer acceptance. And now Seegrid has introduced the world’s first vision-guided autonomous robot.