Cyclic Angles and Value Encoding

cycle_value.hpp and float_encoder.hpp provide two independent numeric utilities: cyclic-angle handling and bounded-range quantization.

CycleValue

#include "cycle_value.hpp"

LibXR::CycleValue<float> heading(-0.2f);
heading += 0.5f;

float normalized = heading.Value();
float error = heading - 0.1f;

The stored value stays in [0, 2π). CycleValue - value returns the shortest signed angular difference in [-π, π), which is useful for cyclic control errors.

Assignment, += and -= renormalize the value. Calculate(value) exposes the normalization operation directly.

FloatEncoder

FloatEncoder<Bits, Scalar> maps [min, max] linearly onto [0, 2^Bits-1]:

#include "float_encoder.hpp"

LibXR::FloatEncoder<12> encoder(-10.0f, 10.0f);
uint32_t code = encoder.Encode(2.5f);
float value = encoder.Decode(code);

Bits ranges from 1 to 31. Encode() clamps to [min, max], scales, then rounds. Quantization step is:

(max - min) / (2^Bits - 1)

Decode() converts the integer code directly back to the floating-point interval. Protocol input should be range-checked to 0..MaxInt() before decoding.

These classes handle values only. Bit packing and byte order belong to the surrounding protocol.