C. Harold Horvat,Matthew Stoffregen
Abstract:The Lonely Runner Conjecture is a number theory problem, dating to 1964. Using dynamical systems theory, we show almost all sets of velocities solve the conjecture. Furthermore, any "traditional" approach of Diophantine approximation cannot solve the problem, and we offer a short list of reformulations of the problem.
What problem does this paper attempt to address?
The problem that this paper attempts to solve is the "Lonely Runner Conjecture". Specifically, this conjecture describes \(k + 1\) runners running on the unit circle. At a certain point in time, the minimum distance of each runner from all other runners is at least \(\frac{1}{k + 1}\).
### The Lonely Runner Conjecture
Consider \(k + 1\) runners running on a unit circle, and the speeds of each runner are \(v_1, v_2,\ldots, v_{k + 1}\), where \(v_i\in\mathbb{R}^+\). Define \(d(a, b)=\|a - b\|_1\), that is, the distance between \(a\) and \(b\).
**Conjecture content**:
There exists a time \(t_i\in\mathbb{R}^+\) such that for any two different runners \(i\) and \(j\), the minimum distance between them satisfies:
\[\min_{i\neq j}d(x_i, x_j)\geq\frac{1}{k + 1}\]
### Main contributions of the paper
The main contribution of this paper lies in proving that the Lonely Runner Conjecture holds for almost all speed vectors \(\mathbf{v}\). Specifically:
1. **Linearly independent speed vectors**:
If the components of the speed vector \(\mathbf{v}\) are linearly independent, then a series of discrete mappings \(T_i:S^1\rightarrow S^1\) can be constructed, defined as:
\[T_i(x)=\|x+\alpha\cdot v_i\|\]
where \(\alpha\in(0,\frac{1}{\max v_i})\setminus\mathbb{Q}\), and \(\alpha\cdot v_i\notin\mathbb{N}\) for all \(i\). These mappings represent irrational rotations on \(S^1\), and their trajectories are dense on \(S^1\).
2. **Density and approximation**:
Since the trajectories of \(T_i\) are dense on \(S^1\), any possible configuration \((x_1, x_2,\ldots, x_k)\) can be approximated arbitrarily well in a finite time. This means that for each runner \(i\), there exists a time \(t = n_i\cdot\alpha\) such that runner \(i\) is lonely.
3. **Excluded point sets**:
The set of excluded points (i.e., points that do not satisfy the conjecture) has a Lebesgue measure of zero. Because any rationally dependent speed vector must be orthogonal to another rational vector, so it lies within an \(n - 1\)-dimensional hyperplane. The set of these hyperplanes has a measure of zero in \(\mathbb{R}^n\).
4. **Coprime points**:
Consider the Lonely Runner Conjecture for \(n + 1\) runners, and assume that the speed of one runner is 0. At this time, only speed vectors of the following form need to be considered:
\[\left(\frac{1}{D_1},\frac{1}{D_2},\ldots,\frac{1}{D_n}\right)\]
where \(D_i\) are pairwise coprime integers, and \(D_i<D_{i + 1}\). By the Chinese Remainder Theorem, a time \(T\) can be found such that the relative positions of each runner satisfy specific conditions, thus proving that the conjecture holds.
### Summary
This paper provides an important progress by proving that the Lonely Runner Conjecture holds for almost all speed vectors. This not only deepens the understanding of this conjecture but also provides new ideas and methods for further research.